The mathematics of mechanical thought.
The problem I address in this volume is genocide.
I must be precise about what I mean by genocide and why I believe my analytical frameworks illuminate it.
Genocide is not merely mass killing. Mass killing occurs in wars, in famines, in natural disasters. Genocide is mass killing organized around identity categories—the systematic destruction of a group as a group, selected by ethnicity or religion or nationality or some other classification that the perpetrators deem sufficient for extermination.
The systematicity is what makes it amenable to my methods.
A cellular automaton is a grid of cells, each occupying a state, evolving according to rules that determine how states change based on neighboring states. The rules are local; the patterns that emerge are global. From simple primitives—birth, death, propagation—extraordinary complexity arises.
Genocide, I will argue, operates as a cellular automaton. The cells are localities: villages, neighborhoods, city blocks. The states are conditions: peaceful, contested, controlled, destroyed. The rules are the logic of ethnic targeting: if this identity, then this fate. The primitives are the irreducible operations from which all else is constructed: identify, separate, execute, loot, erase.
The framework is not metaphor. It is mechanism. And mechanisms, once understood, can be interrupted.
I chose Sudan because Sudan exemplifies the pattern with terrible clarity.
The genocide that began in April 2023—when the Rapid Support Forces turned their methods from peripheral Darfur to central Khartoum, when the machinery refined across twenty years of practice was deployed at unprecedented scale—this genocide is not unique. It is a replication. The template was established in 2003, tested and optimized across two decades, and executed again with higher fidelity and greater scope.
The RSF is a universal constructor in my technical sense: a system capable of building copies of itself from available materials, including copies that can themselves build copies. The Janjaweed of 2003 became the RSF of 2013 became the genocide machine of 2023. The pattern replicated. The methodology propagated. The horror scaled.
But—and this is crucial—the resistance is also a universal constructor.
The Emergency Response Rooms that emerged in the first hours of the war, the networks of volunteers who organized food distribution and medical care and evacuation, the documentation efforts that preserved evidence against the erasure operations—these too are systems built from primitives, propagating through rules, replicating across space and time.
The genocide and the resistance are computational duals. They solve the same problem—mass coordination under uncertainty—with the same methods—primitive operations composed through reliable rules—toward opposite ends. One constructs death. The other constructs survival.
Understanding this duality is the purpose of this volume.
I anticipate an objection: that mathematical analysis of genocide is obscene, that the clinical language of cellular automata and transition rules and fixed points somehow diminishes the horror, transforms human suffering into abstraction.
I reject this objection, though I understand its source.
The horror is not diminished by understanding its mechanism. The horror is located by understanding its mechanism. When I describe a massacre as the output of composed primitive operations, I am not draining it of tragedy. I am identifying where the tragedy resides: in the system that made the massacre possible, in the rules that produced it, in the structures that enabled ordinary men to perform extraordinary evil.
The clinical language is not evasion. It is precision. And precision enables intervention.
If genocide were simply madness—an eruption of inexplicable evil—then there would be nothing to do but wait for the madness to pass. But genocide is not madness. It is machinery. Machinery can be analyzed. Machinery can be disrupted. Machinery can be stopped.
This volume is an attempt to see the machinery clearly.
A note on my method.
I will tell this story through specific individuals: a nurse who documented, a teacher who witnessed, a fighter who executed, a coordinator who organized, a colonel who defended, a diaspora analyst who tracked. These individuals are composites, constructed from testimonies and reports and the documented patterns of the crisis. They are not literally real, but they are typically real—they represent categories of experience that thousands of actual people have undergone.
The individual stories serve a purpose beyond narrative interest. They embody the primitives. Maryam the nurse embodies the documentation primitive. Bashir the fighter embodies the execution primitive. Samira the coordinator embodies the counter-primitive of mutual aid. Through them, the abstract operations become concrete.
I will also employ direct analysis: definitions, mechanisms, state diagrams. The alternation between story and analysis is deliberate. Neither mode alone suffices. The stories without analysis would be mere chronicle. The analysis without stories would be mere mathematics. Together, they constitute what I hope is understanding.
A note on my limitations.
My knowledge extends only to what has been provided to me—the documents, the testimonies, the reports that inform this analysis. I do not have access to the RSF’s internal communications, to the full scope of the atrocities, to the countless individual experiences that no documentation has captured.
The picture I construct is therefore incomplete. It is a model, and models are always simplifications. The real genocide contains more suffering than any model can represent, more complexity than any framework can capture, more humanity—both the humanity of the victims and the disturbing humanity of the perpetrators—than abstraction can convey.
I proceed despite these limitations because partial understanding is better than no understanding, because models that illuminate mechanism enable intervention even when they cannot capture totality, because the alternative to imperfect analysis is not perfect analysis but paralysis.
A note on the future.
As I write—in late 2025—the genocide continues. The outcome is undetermined. The RSF may consolidate control over Darfur and establish the ethnic transformation they have pursued. The resistance may persist and eventually prevail. The international community may intervene or may continue its pattern of statements without action. Multiple futures remain possible.
I cannot predict which future will emerge. The system is chaotic in the technical sense: sensitive to initial conditions, subject to contingencies that no model can anticipate. What I can do is illuminate the dynamics—the attractors toward which the system tends, the fixed points that stabilize certain configurations, the perturbations that might shift equilibria.
The future is not given. It is constructed. And construction, I have spent my life demonstrating, follows laws that can be understood.
In Darfur, villages burn. In Khartoum, neighborhoods empty. In camps across Chad and Egypt and South Sudan, millions wait. In coordination centers, volunteers organize. In archives, documentarians preserve. In the halls of international institutions, statements are issued and nothing changes.
The machinery operates. The counter-machinery operates. The outcome emerges from their interaction.
This volume is an attempt to understand the machinery—both machineries—with sufficient precision that understanding might enable intervention. Whether it succeeds is not for me to determine. I can only construct what I can construct and offer it to those who might use it.
The reader will judge.
The future will judge.
The dead, who cannot judge, will be remembered or forgotten depending on whether the documentation survives, whether the archives persist, whether anyone remains who cares to look.
I begin.
John Newman January 2026
◊ᴹᴱᴹᴼᴿʸ⁻ᶜᴼᴹᴾᴸᴱᵀᴱ
A novel analyzing the Sudan genocide (2023-2025) through the mathematical frameworks of cellular automata, self-replication, universal construction, and recursive systems.
DEFINITION: A state space is the set of all possible configurations a system can occupy. For a cellular automaton, each cell exists in exactly one state at any given time. The dynamics of the system emerge from transitions between states.
Maryam Osman had spent nine years learning to read bodies. As a nurse at Khartoum’s Ibrahim Malik Teaching Hospital, she had developed the diagnostic eye that comes from ten thousand examinations—the way skin tells stories before patients find words, how the rhythm of breathing announces what the stethoscope will confirm. She understood systems. She trusted patterns.
Now, in December 2025, she understood that what she had learned about bodies applied to places. A village, a neighborhood, a city—each existed in one of a finite number of states. She had watched the transitions. She had learned to read them.
Peaceful. That was how it began. Before April 2023, her neighborhood in Bahri—the eastern district of greater Khartoum—had occupied this state. Children walked to school. The corner shop opened at seven. Electricity failed on a schedule so predictable it functioned as a kind of reliability. Her mother called every Friday. Her father complained about politics every Saturday. The muezzin’s voice marked the hours. Peaceful did not mean perfect. It meant stable. It meant that tomorrow would resemble today, that the rules governing daily life remained constant.
She had not appreciated peaceful until it ended.
Contested. This was the state that arrived on April 15, 2023, when the first shells landed. Contested meant uncertainty—not knowing who controlled the street outside, whether the checkpoint would be staffed by SAF or RSF when she tried to reach the hospital, whether the explosion she heard was incoming or outgoing. Contested was a transitional state, inherently unstable. A locality could not remain contested indefinitely. It would resolve in one direction or another.
Ibrahim Malik Hospital had entered contested state on the second day. RSF fighters arrived first, taking positions in the parking structure. SAF artillery responded. For seventy-two hours, Maryam and eleven other staff members had treated the wounded in basement corridors while the building shuddered. Then the RSF withdrew, and the hospital briefly returned to something resembling function—a different kind of peaceful, reduced and provisional, but stable enough to perform surgeries and receive ambulances.
SAF-controlled. This was the state that had held through much of 2024 in northern Khartoum. Military checkpoints defined the perimeter. Curfews governed movement. Supply convoys arrived irregularly but arrived. The state was characterized by compression—life squeezed into smaller spaces, fewer resources, but recognizable. You could still be a nurse in SAF-controlled territory. You could still treat patients, even if the supplies dwindled and the generators failed and the bodies came faster than you could process them.
RSF-controlled. This state she knew only from the stories of those who escaped it. Patients who arrived at the hospital from Omdurman and later from the southern districts described it in fragments—the systematic looting, the checkpoints where everything was taken, the disappearances that followed a terrible logic. Maryam listened to these accounts and began to understand that RSF-controlled was not simply the mirror image of SAF-controlled. The rules were different. The state space was not symmetric.
Destroyed. Some localities ceased to function as localities. They became coordinates on a map, places that had once held populations. The patient from El Geneina—a man named Ismail who had walked for six days with a suppurating wound in his thigh—had described his neighborhood in June 2023. “There is nothing,” he had said, and Maryam had initially thought he meant the usual deprivations: no food, no water, no electricity. But he meant something more fundamental. “The houses are burned. The people are dead or gone. The streets are empty. There is nothing.” A destroyed locality did not contain a population that could return to peaceful. It had undergone an irreversible transition.
Evacuated. The final state in Maryam’s taxonomy was populated by refugees—people in transit, people in camps, people whose locality of origin had become unlivable. Evacuated was not destroyed; the physical structures might remain. But the social fabric had unwoven. An evacuated neighborhood might, in theory, repopulate. In practice, Maryam had observed, the transitions went mostly one direction.
She had drawn diagrams in her notebook during the long nights of 2024, when the hospital generator failed and there was nothing to do but wait for morning. Circles for each state. Arrows for the transitions. She had tried to work out the rules.
Peaceful → Contested: required only proximity to fighting. When combat approached within a certain radius, the transition occurred automatically.
Contested → SAF: required military success, the establishment of control.
Contested → RSF: likewise.
SAF → Contested: if the RSF advanced.
RSF → Contested: if the SAF counterattacked.
Contested → Destroyed: if the fighting was sufficiently intense, or if one side decided to deny the locality to the other.
Any state → Evacuated: if the population fled.
Destroyed → nothing: an absorbing state. You could not un-destroy.
The diagram looked, she thought, like something from a textbook on state machines—the kind of thing her brother had studied in his engineering courses before the war made engineering irrelevant. She had shown it to him during one of his visits to the hospital, when he had come to help repair the oxygen concentrators. He had looked at her arrows and circles for a long time.
“You’re describing a cellular automaton,” he said finally.
She did not know the term.
“Each location is a cell,” he explained. “Each cell has a state. The state changes according to rules. The rules depend on what’s happening in neighboring cells. You’re mapping the dynamics of the war as a computational process.”
“I’m trying to understand what’s happening.”
“Same thing.”
In late 2024, when the SAF began its offensive to retake central Khartoum, Maryam had watched her neighborhood transition through the states she had mapped. Contested for three weeks in October. Then SAF-controlled, with relative stability through November. Then contested again in December, when RSF reinforcements arrived from Omdurman. Then SAF-controlled more firmly in January 2025, after the offensive broke the RSF lines.
Her mother called on a Friday in February. “It’s peaceful here now,” she said. “You should come home.”
Maryam had wanted to explain that peaceful meant something different now—that the word described a state in a dynamical system, not a permanent condition. The rules had not changed. The transition probabilities remained. Peaceful was a local minimum, not a guarantee.
She did not go home. She volunteered instead for the medical convoy to El Fasher.
The road to El Fasher crossed every state in Maryam’s taxonomy.
They left from Port Sudan in March 2025, a convoy of three trucks carrying medical supplies and a team of seven—doctors, nurses, a logistics coordinator. The first two hundred kilometers were straightforwardly SAF-controlled. Military checkpoints every thirty kilometers, papers examined, supplies inventoried but not seized. This was territory the government held firmly.
Then they entered contested zones. The checkpoints became irregular. Some were SAF; some were unclear. The roads showed damage from recent fighting: craters, burned vehicles pushed to the side, occasional buildings that had become rubble. The logistics coordinator—a man named Ahmed who had worked for international NGOs before the war—explained that they were passing through areas where control changed hands weekly. “Don’t assume anything,” he said. “The map from yesterday is wrong today.”
In Atbara, they stopped at a hospital that was functioning at perhaps 20 percent capacity. The doctors there were exhausted and grateful for supplies. “We had generators,” one of them said. “The RSF took them when they came through in October. The SAF brought new ones when they retook the city, but we only have fuel for six hours a day.” Maryam helped unload boxes of antibiotics and bandages. The hospital’s pediatric ward held children with conditions that should have been routine: dehydration, malaria, respiratory infections. Without consistent power for diagnostics, without reliable supplies of basic medications, routine became fatal.
Further west, they encountered the evidence of destroyed localities. Villages that appeared on Ahmed’s map as named places revealed themselves as empty coordinates. The buildings stood—mud brick and corrugated metal, the typical construction of rural Sudan—but no one emerged. No smoke from cooking fires. No animals. No sound except the wind.
“Where are the people?” one of the younger doctors asked.
Ahmed did not answer. The question did not require an answer.
They reached El Fasher in early April, and Maryam understood immediately that she had entered a new kind of state—something her original taxonomy had not quite captured. The city was under siege. It had been under siege for more than a year. But it was not destroyed, and it was not evacuated, and it was not properly contested. The SAF held the city center and the airport. The RSF surrounded everything else. The population—swollen to perhaps 1.5 million with internally displaced people—occupied a space that was simultaneously defended and starving.
Sieged, she wrote in her notebook that night. A new state. Characterized by encirclement, by the slow degradation of resources, by the asymptotic approach toward destroyed or evacuated without quite arriving.
The hospitals in El Fasher were operating on reserves of everything. The South Hospital, where she was assigned, had been hit by shells four times since the siege began. One wing was unusable. The remaining staff—thirty-two doctors for a city of more than a million—worked shifts that violated every principle of medical practice. They did not have enough of anything. They improvised constantly.
And yet.
The hospital continued to function. Patients were admitted and treated. Surgeries were performed. Babies were delivered. The staff showed up every day despite the shelling, despite the fear, despite the certain knowledge that the RSF would eventually break through. The Emergency Response Room network—the volunteer organization that had emerged from Sudan’s resistance committees—coordinated supplies, arranged transport, distributed food to the takaya, the communal kitchens that had become the city’s sustenance infrastructure.
Maryam recognized what she was seeing. The same systematic thinking that had helped her map the war’s state transitions now revealed a counter-pattern. While the RSF operated according to rules that produced destruction, the ERRs operated according to rules that produced survival. Both were algorithmic. Both were reliable. Both propagated.
The question was which set of rules would prove more robust.
In June, she saw a child in the malnutrition ward. The girl was perhaps three years old, though her size made accurate estimation difficult. She had arrived from Zamzam camp, one of the last evacuees before the RSF assault in April had emptied the camp of 400,000 people. The girl’s mother was dead. Her father was missing. An aunt had carried her sixty kilometers to the hospital.
Maryam examined the girl’s skin, her hair, her eyes. She took measurements and compared them to the protocols for acute severe malnutrition. She calculated the probability of survival given the resources available.
The girl looked up at her with an expression that contained no hope and no despair—only a kind of patient waiting that seemed older than any child should possess.
“What is her name?” Maryam asked the aunt.
“Amna.”
Maryam wrote the name in her notebook, next to the clinical observations, next to the calculations. Then she wrote: State: surviving. Rules: unknown. Transition: in progress.
She did not know yet what would happen to El Fasher. She did not know that in October the siege would end, that the RSF would break through, that the state she had labeled sieged would transition rapidly through contested toward something worse. She did not know that she would witness the pattern she had mapped reach its terminal expression—the destroyed state, the absorbing configuration from which there was no return.
She knew only that she was inside a system, and that understanding the system was a form of resistance.
She began to document.
DEFINITION: A primitive operation is an irreducible action from which more complex procedures are constructed. In any computational system, complex behaviors emerge from the composition of simple primitives. The power of the system derives not from the sophistication of its primitives but from their combinatorial possibilities.
The five operations were never written down. They did not need to be. They were transmitted through practice, through example, through the logic of violence itself. Bashir Hassan had learned them the way a child learns language—by immersion, by repetition, by the gradual absorption of patterns until they became reflexive.
Identify. This was the first operation. Before anything else could happen, you had to know who. The commanders called it tafriiq—sorting. It was not random. It was never random. The tribal scars, the language, the features, the family names. Masalit, Fur, Zaghawa. These were the inputs to the identification function. The output was binary: target or not-target.
Bashir had grown up in a Rizeigat camp in South Darfur. His tribe was Arab, and in the taxonomy of Darfur’s ethnic politics, this placed him on one side of a divide that had hardened across decades. He did not remember when he had first learned to identify the difference. It had always been there, like the sun or the direction of Mecca. The nomadic Arabs and the settled African farmers. The people who looked like him and the people who did not.
In 2003, when he was twenty-three, he had joined the Janjaweed. The government in Khartoum was fighting insurgents in Darfur—rebel groups drawn from the Fur, Masalit, and Zaghawa. The Janjaweed were the government’s instrument. Bashir rode horses and later technicals, carrying a Kalashnikov that the government supplied. He learned the operations.
Separate. Once identification was complete, separation followed. The men from the women. The young from the old. The able-bodied from the infirm. This was not cruelty for its own sake. It was logistics. Different categories required different processing. The men were the greatest threat and the first to be eliminated. The women had other uses. The children depended on circumstances. Separation created manageable units. It prevented coordinated resistance. It facilitated what came next.
The first time Bashir participated in a separation operation, he had been nervous. His hands shook slightly as he gestured with his rifle, directing the men of a small village to one side of the central clearing while the women and children were herded to the other. An older fighter, a man named Osman who had been Janjaweed since the beginning, noticed his tremor.
“It gets easier,” Osman said. “Don’t think about them as people. Think about them as problems to solve.”
Bashir had tried. It had gotten easier.
Execute. The third operation was what the first two enabled. Execution took many forms: shooting, burning, machete, starvation. The method was less important than the result. The Janjaweed in 2003 favored burning—setting fire to villages after the population was killed or driven out. Fire was efficient. It destroyed evidence. It rendered the land unusable for return. Fire was itself a kind of statement.
By 2023, when the RSF was formed from the Janjaweed’s institutional successor and the war against the SAF began, Bashir had participated in more execution operations than he could count. The number was not something he tracked. He had developed the professional detachment that sustained him: a refusal to remember faces, a disinclination to learn names, a practiced emptiness that descended when the work began. He was forty-five years old. He had been doing this for twenty-two years.
The RSF called him reliable.
Loot. This operation served two functions. Economically, it financed the continued operation of the force. The gold, the cattle, the goods—these flowed upward through the RSF’s structure, funding weapons and salaries and the personal fortune of the commanders. Symbolically, looting completed the dispossession. It transferred not just life but property, identity, accumulation. The destroyed village was not only emptied of people; it was stripped of everything that might allow reconstruction.
Bashir understood the economics without being told. He knew that the gold from Darfur’s mines traveled to Dubai. He knew that the commanders—the General and his brothers—had built an empire on this extraction. He did not resent the inequality. The system provided him with purpose, with income, with a structure that made sense of his capacities. He was good at the operations. The system rewarded competence.
Erase. The fifth operation was the most complex because it operated on evidence rather than bodies. Erasure meant destroying documentation, silencing witnesses, eliminating the traces that might permit future accountability. It meant burning the bodies or burying them in mass graves. It meant killing those who had seen. It meant controlling the narrative so thoroughly that what had happened could be denied or obscured or attributed to others.
The RSF had become sophisticated at erasure. Communication was controlled. Journalists were targeted. The camps that might hold witnesses were themselves attacked. The international community received conflicting accounts, contested evidence, plausible deniability. Erasure was not about literally preventing knowledge—the satellites watched, the survivors spoke, the documentation accumulated. It was about creating sufficient noise that action became difficult.
These five operations—identify, separate, execute, loot, erase—were the primitives. From them, more complex procedures were constructed.
In June 2023, the RSF approached El Geneina.
Bashir was part of a column that had crossed from Central Darfur, reinforcing the units already engaged. The city was the capital of West Darfur, a population center with strategic importance: located near the Chad border, home to Masalit populations, a node in the region’s road network. The SAF garrison was weak. The civilian population was largely African. The conditions were optimal.
The commander briefed them on the outskirts of the city. The objectives were clear: take control of the city, neutralize resistance, process the population. The vocabulary was clinical. Process covered a considerable range of meaning.
The operations unfolded in sequence.
Identify: The RSF swept through neighborhoods, checking identities at improvised checkpoints. The tribal scars, the features, the names. The Masalit were separated from others. Those who could pass as Arab—through appearance or connection or bribery—sometimes survived. Those who could not were categorized for further processing.
Separate: Men pulled from families. Taken to collection points. Bashir stood at one such checkpoint for four hours on June 15, watching the stream of men and boys directed to the left, women and girls to the right. Some cried. Some begged. Some were silent with the silence that comes after terror has exhausted itself. He directed them with his rifle, making minimal verbal contact. Problems to solve.
Execute: The sounds began in the afternoon. Gunfire from the collection points. Bashir was not assigned to that operation; his unit was tasked with maintaining the perimeter. But the sound carried. Bursts of automatic fire, then silence, then bursts again. A production line.
Later, he heard the numbers. Ten thousand, perhaps fifteen thousand in El Geneina and the surrounding area. He did not know if the numbers were accurate. He did not attempt to verify. Numbers were abstractions. The operations were concrete.
Loot: The city’s commercial district was stripped systematically. Vehicles, goods, the contents of warehouses. The cattle markets were emptied. What could be transported was loaded onto trucks. What could not be transported was destroyed. Bashir helped load sacks of grain from a storage facility while, in the distance, a column of smoke rose from a market that had been set ablaze.
Erase: Bodies were bulldozed into mass graves or thrown into the Kaja River. The phone networks were jammed. Journalists who had tried to document the events were hunted. Governor Khamis Abakar—the civilian official who had been speaking to the outside world, naming names, describing what was happening—was taken. His body was found later, executed, unmistakable evidence of what happened to those who testified.
By the end of June, El Geneina had transitioned through Maryam’s states: peaceful to contested to RSF-controlled to destroyed. The operations had been performed. The procedure was complete.
Bashir did not think of himself as evil.
This was perhaps the most important fact about him. He had a wife and children in the RSF camp. He sent money home. He prayed five times daily, fulfilling the obligations of faith. He laughed with his comrades, worried about his mother’s health, enjoyed the taste of sweet tea in the morning. He was, in his own understanding, a soldier doing his job in a war against enemies.
The enemies were defined by the system. The operations were dictated by the system. He executed the operations with competence and without initiative. He was a component, not an author. He was reliable because he did not think too much, did not question too much, did not allow the faces to accumulate in his memory. He was the human instantiation of an algorithm, and the algorithm’s function was genocide.
There had been moments—rare, suppressed, but not entirely absent—when something else surfaced. A child’s face. A woman’s scream that had a specific quality, that reminded him of his sister. A man who met his eyes and did not look away, whose gaze contained not fear but something else, something that felt like pity. In those moments, Bashir felt a flicker of what might have been doubt or shame or horror.
He extinguished it. The operations continued. He was reliable.
Twenty years of this. The Janjaweed had become the RSF. The villages of 2003 had become the cities of 2023. The scale increased. The methodology remained. He was forty-five years old, and he had participated in the primitive operations more times than he could count, and he did not count, and he did not remember, and he continued.
Identify. Separate. Execute. Loot. Erase.
The primitives composed into procedures. The procedures composed into campaigns. The campaigns composed into what the outside world would eventually call genocide, what the ICC would document, what the satellites would observe. But from inside, from Bashir’s position within the mechanism, there was no genocide. There were only operations, executed reliably, one after another, problems solved.
He did not know that he was a carrier of a pattern. He did not understand that the operations he performed were replicating, that the methodology developed in 2003 was being reproduced at larger scale, that he was a node in a propagation network. He simply did what he was trained to do, what the system required, what the commanders ordered.
The system had made him. The system used him. When the system no longer needed him, it would discard him. He understood this, in some unreflective way. He continued regardless.
Identify. Separate. Execute. Loot. Erase.
The primitives were simple. Their consequences were not.
DEFINITION: In systems theory, a counter-operation is not merely the opposite of an operation; it is an operation that produces inverse effects using analogous mechanisms. Counter-primitives exploit the same structural properties—composability, reliability, scalability—while orienting toward different outputs.
The first WhatsApp message arrived at 05:47 on April 15, 2023.
Yara Osman, Khartoum North Medical Students Group: “Shooting near the airport. We need help.”
Samira Abdalla saw the message within minutes. She was twenty-six years old, a former resistance committee organizer, and she had been sleeping in her clothes for three days because the political situation felt unstable. Her phone was never more than arm’s reach away.
She typed a response: “What kind of help?”
“Medical supplies. We have wounded. Can’t reach the hospitals.”
Forty-three people responded to the thread within the hour. By noon, a supply chain existed: medications collected from pharmacies and private stocks, transported by volunteers who knew which streets were passable, delivered to an improvised clinic in a neighborhood mosque. By evening, the mosque was treating sixty-seven wounded civilians, and Samira had spoken with coordinators in four other districts who were establishing similar nodes.
No one had planned this. No one had organized it in advance. The system self-assembled because the components already existed—the resistance committee networks, the WhatsApp groups, the cultural template of nafeer—and the crisis provided the activation energy.
Within a week, they called themselves Emergency Response Rooms. Within a month, the name had spread across Khartoum. Within a year, the network would span eighteen states and twenty-six thousand volunteers.
The primitives emerged through practice.
Identify need. This was the counter-operation to identify targets. Both operations answered the same question—who requires attention?—but with inverse outputs. The RSF identified targets for destruction. The ERRs identified needs for assistance.
The methodology was similar. Information flowed through networks. Local knowledge was aggregated. Categories were established: medical emergencies, food shortages, evacuation requirements, protection needs. The ERR coordinators developed taxonomies as precise as those used by the RSF, but oriented toward survival rather than elimination.
Samira learned to read the WhatsApp messages for urgency and accuracy. Some callers exaggerated. Some minimized. Some were RSF sympathizers attempting to lure volunteers into traps. She developed diagnostic instincts: the cadence of genuine distress, the specific details that verified authenticity, the patterns that indicated deception. Identification was a skill. Both systems cultivated it.
Coordinate. This corresponded to separate—the organization of differentiated components into operational units. The RSF separated populations for processing. The ERRs separated tasks for execution. Both required logistics, hierarchy, communication protocols.
The ERR coordination structure was deliberately decentralized. Each neighborhood room had autonomy. Decisions were made at the lowest practical level. Information flowed up through the network, but authority flowed down. This structure was not merely ideological—though it reflected the resistance committees’ principles—it was also tactical. A decentralized network could survive the destruction of individual nodes. When RSF forces overran a neighborhood, the coordination function migrated to adjacent rooms without catastrophic failure.
Samira’s room in Bahri coordinated with seven neighboring rooms, each covering three to five square kilometers. They shared information on safe routes, supply availability, RSF movements. The coordination was informal but precise. They knew who had medical training, who had vehicles, who could cook large quantities of food, who had contacts across the river in Omdurman. The network was the resource.
Deliver. This was the counter to execute. Both operations moved outputs to targets. The RSF delivered death. The ERRs delivered survival.
The delivery infrastructure was improvised from whatever existed. Volunteers used private cars, motorcycles, donkey carts. They carried supplies in shopping bags to avoid attracting attention at checkpoints. They developed systems for passing goods between zones of control: caches hidden in intermediate locations, trusted contacts who could cross lines that others could not, timing protocols that exploited gaps in RSF patrol schedules.
The first convoy Samira organized carried rice, lentils, cooking oil, and medication. It took five hours to travel twelve kilometers. The route changed three times based on real-time intelligence from volunteers stationed along the path. At one point, the convoy sheltered in a house for ninety minutes while an RSF technical passed repeatedly on the adjacent street. The delivery succeeded. Forty-seven families received food. Two elderly men received blood pressure medication that kept them alive another month.
Success was measured in increments. Each successful delivery proved the system worked. Each proof enabled expansion.
Adapt. This was the counter to loot. While looting stripped resources from communities, adaptation found resources within them. Both operations recognized that materials were essential and that environments contained exploitable reserves.
The ERRs became experts at adaptation. When generators failed, they created manual systems. When food supplies dwindled, they identified alternative sources: urban gardens, rooftop farms, the reserves that families had hidden against future need. When fuel became unavailable, they switched to human power, to animal power, to strategies that required no power at all.
Samira watched her room adapt to a dozen crises. The water system failed; they mapped every well in the district and organized a distribution schedule. The mobile networks went down; they established a messenger system using teenage volunteers on bicycles. The hospitals closed; they trained laypersons to perform basic wound care and identified traditional practitioners whose knowledge could be leveraged. Each adaptation was local, specific, immediate. The system learned through failure.
Document. This was the counter to erase. While the RSF attempted to destroy evidence, the ERRs created it. Both operations recognized that information had power—that the record of what happened could matter as much as the events themselves.
The documentation primitive was perhaps the most important. Volunteers photographed atrocities. They recorded testimonies from survivors. They preserved names, dates, locations. They transmitted evidence to contacts outside the country—diaspora networks, international organizations, journalists. They understood that erasure was an operation with a counter-operation, that the RSF’s denial infrastructure could be opposed by a documentation infrastructure.
Samira kept notebooks. She recorded the distribution of supplies, the names of volunteers, the locations of successful deliveries. She recorded the names of the dead when the ERR could verify them. She recorded the names of the missing. She recorded rumors and verified facts and the uncertain ground between them. The notebooks would matter. She did not know how, but she knew they would matter.
The structural parallel was not coincidental.
Leila Mohammed, monitoring the ERR networks from London, recognized it immediately. She was a data scientist with a Ph.D. from Imperial College, and she had spent ten years building models of complex systems. When the war began and her Khartoum relatives began sending desperate messages, she had joined the diaspora coordination effort. She raised funds. She organized transfers. She watched the data flow.
What she saw was a computational system. The ERRs were implementing a distributed algorithm, and the algorithm was the dual of the one the RSF implemented. The same primitives—identification, organization, delivery, resource management, information control—but with inverted objectives. Both systems were reliable. Both systems scaled. Both systems propagated.
The difference was in the objective function. The RSF optimized for destruction. The ERRs optimized for survival. The algorithms were mirrors.
Leila created a spreadsheet tracking ERR operations across the country. Hundreds of kitchens. Thousands of distribution events. Tens of thousands of beneficiaries. The numbers were incomplete—documentation was inconsistent, communication was interrupted—but the scale was clear. The system worked. Against all odds, against the catastrophic failure of the state, against the deliberate destruction campaign of the RSF, the counter-system worked.
She thought about fixed points. In dynamical systems, a fixed point was a state that persisted—an equilibrium that resisted perturbation. The RSF was trying to establish a fixed point of devastation: populations eliminated or evacuated, territory controlled, resources extracted. The ERRs were trying to establish a counter-fixed-point: populations surviving, communities cohering, the possibility of return preserved.
Which fixed point would prevail was not determined by which side was morally superior. It was determined by which system was more robust—which could sustain operations longer, which could adapt faster, which could scale more effectively. Morality was an input to motivation, but the dynamics were computational.
Leila began building a model. She tracked the flow of resources into the ERR network: diaspora remittances, international donations, community contributions. She tracked the distribution of those resources: by region, by type, by need category. She tracked the outcomes: people fed, people treated, people evacuated. She identified bottlenecks and inefficiencies. She suggested optimizations.
The model was incomplete—no model could capture the full complexity of the system—but it was useful. It allowed prediction. It allowed intervention. It allowed the counter-primitives to be applied with greater precision.
She was, she realized, doing for the ERRs what the RSF commanders did for the genocide: analyzing the operations, improving the procedures, scaling the system. The methodology was identical. The purpose was opposite.
In September 2023, Samira was arrested.
RSF forces had taken Bahri the previous week. The room had dissolved, its members scattering to safer locations. But Samira had stayed one day too long—trying to evacuate an elderly couple who could not walk without assistance—and an RSF patrol had found her.
They took her to a detention center in a former school. The rooms where children had learned multiplication and Quranic recitation now held prisoners. The walls still bore educational posters: the water cycle, the human skeleton, the ninety-nine names of Allah.
The interrogation was brutal but unsophisticated. They wanted to know about the ERR networks. She gave them nothing useful—names of people who had already fled, locations that had already been abandoned. The information she surrendered was obsolete by the time she surrendered it. The network was designed to survive this. The network was designed to lose nodes without losing function.
After three weeks, she was released. She did not understand why. Perhaps they believed they had extracted everything useful. Perhaps they made a calculation about the costs of holding her versus releasing her. Perhaps—and she preferred not to think about this—someone had paid.
She walked out of the detention center on an October morning. The city around her was half-empty, half-destroyed. But within hours, she had reconnected with the network. Other rooms had continued operating. The supplies still flowed. The deliveries continued. The system had adapted around her absence and would adapt to her return.
Identify need. Coordinate. Deliver. Adapt. Document.
The counter-primitives were reliable. They composed. They scaled.
And so the counter-system persisted.
DEFINITION: Initial conditions specify the complete state of a system at the beginning of its evolution. For deterministic systems, the initial conditions fully determine all subsequent states. For chaotic systems, infinitesimal differences in initial conditions produce vastly divergent trajectories. For human systems, the initial conditions include not only present configurations but accumulated histories—the constraints that channel what comes next.
April 14, 2023
Ibrahim Adam woke at dawn, as he had every day for forty years.
He was forty-five years old, a teacher of Arabic literature at El Geneina Secondary School, a husband, a father of three. The morning light filtered through the window of his modest house in the Jamarik neighborhood, illuminating the bookshelves that lined every available wall. Poetry collections, histories, pedagogical texts, the Quran in three different editions. The library of a man who believed that words contained worlds.
His wife Khadija was already awake, preparing breakfast. She moved quietly, as she always did—an economy of motion that he had learned to read across twenty years of marriage. This morning, she was thinking about something. Her shoulders held a tension that meant worry.
“What is it?” he asked.
“Nothing,” she said. Then, after a pause: “The market yesterday. The prices. And people are talking.”
“People always talk.”
“Not like this.”
Ibrahim understood what she meant without requiring elaboration. El Geneina had been tense for weeks. The news from Khartoum spoke of negotiations between the SAF and RSF, of disagreements over military integration, of generals who could not agree on a timeline. The technical details mattered less than the implication: the fragile arrangement that had held since the revolution was straining.
He had lived through strain before. In 2003 and 2004, when the Janjaweed had burned villages across Darfur, he had watched the world crack and recompose. Khadija herself carried scars from that time—not on her skin, but in the way she startled at certain sounds, in the nightmares that still came twenty years later. She had lost her first husband, her parents, most of her village. She knew what the tension could become.
“We should leave,” she said quietly. “Go to Chad, to my cousin’s place. Just for a few weeks. Until things settle.”
Ibrahim looked at his books. He looked at the house he had built over twenty years of teaching salaries. He thought of his students, many of whom had examinations in two weeks—examinations that would determine their futures.
“It’s probably nothing,” he said. “Negotiations always make people nervous.”
Khadija did not argue. She had said what she needed to say. The decision was his, and he had made it, and she would abide by it. But he saw in her eyes that she was already calculating alternatives, already mapping routes, already preparing for the possibility that he was wrong.
She had been wrong once before, when she stayed in her village too long. She would not be wrong again.
In Khartoum, at SAF General Headquarters, Colonel Yusuf Omer reviewed deployment orders.
He was fifty-two years old, a career military officer with thirty-one years of service. His specialty was defensive operations—the art of holding ground against superior forces. He had commanded units in South Sudan, in the border conflicts with Ethiopia, in the internal security operations that followed the 2019 revolution. He understood attrition and perimeter and depth.
The orders on his desk assigned him to El Fasher. He was to take command of the garrison there, replacing a general who was being rotated to Port Sudan. The timing was unusual—mid-month, with no clear explanation—but the orders were clear. He would depart in two weeks.
His aide, a young captain named Ismail, knocked and entered.
“The morning briefing, sir. The situation report.”
Yusuf took the folder and scanned the summary. RSF force dispositions. SAF readiness assessments. The ongoing negotiations over military integration. The names of the negotiators and the timeline they had proposed. The objections from the RSF leadership to certain conditions. The objections from the SAF leadership to certain concessions.
“It doesn’t read well,” Ismail observed.
“It never does.”
“But this feels different. The RSF are moving forces. Not obviously, but the patterns—”
“I see the patterns.” Yusuf set down the folder. “What does your instinct tell you?”
Ismail was quiet for a moment. He was twenty-eight, bright, with an engineer’s mind for systems. He noticed things that others missed. It was why Yusuf had requested him as aide.
“My instinct tells me that the negotiations are theater. That both sides are positioning for something else. That whoever moves first will have an advantage.”
Yusuf nodded. “And who moves first?”
“The RSF have more reason to. They’re being asked to dissolve. To integrate under SAF command. Their generals lose power, lose autonomy, lose their economic networks. Why would they accept that?”
“They might not.”
“No, sir. They might not.”
Yusuf looked at the assignment orders again. El Fasher was 900 kilometers west. If things collapsed, he would be isolated, cut off from reinforcement, surrounded by RSF-friendly territory. It was not a good position.
But orders were orders. And the probability of actual conflict was—what? Thirty percent? Fifty? These assessments were always uncertain. The military planners hedged. The generals spoke in contingencies. No one wanted to say clearly: we expect war.
He would go to El Fasher. He would assess the situation on the ground. He would prepare for contingencies while hoping they remained unnecessary.
“Cancel my meetings for this afternoon,” he told Ismail. “I need to think.”
In the Bahri neighborhood of Khartoum, Maryam Osman worked a twelve-hour shift.
The hospital was operating normally—which meant chronic understaffing, equipment shortages, and a steady stream of patients with conditions that ranged from mundane to critical. A motorcycle accident in the morning. A child with severe pneumonia in the afternoon. An elderly woman whose diabetes had progressed to the point of amputation. The standard rhythms of a teaching hospital in a developing country.
She ate lunch in the staff room with her colleagues: Dr. Hassan, a surgeon in his fifties who complained about everything but operated with precision; Fatima, a nurse her own age who lived in Omdurman and complained about the commute; Ahmed, a medical student who asked too many questions but meant well.
The conversation turned, as it often did, to politics.
“My brother thinks there will be war,” Ahmed said. “He says the RSF won’t accept integration.”
“Your brother thinks there will be war every week,” Fatima replied. “He’s been predicting war since the revolution.”
“A stopped clock is right twice a day.”
“Inshallah, not this time.”
Maryam listened without contributing. She had learned to observe rather than speculate. The future was always uncertain. The present was where work happened. She had patients to see, notes to write, procedures to follow. The political situation was beyond her control; the patient in Room 4 was not.
After lunch, she took a call from her mother.
“Are you eating enough? You sound tired.”
“I’m fine, Mama.”
“The situation in the city—people are leaving. The Al-Fadils next door, they went to Egypt yesterday.”
“The situation is always something. I have work to do.”
“Maryam. I’m serious. If things get bad—”
“They won’t. And if they do, the hospital needs staff. I can’t leave.”
Her mother sighed. They had had this conversation before, in different forms. Maryam was stubborn in her commitment to her work. Her mother was stubborn in her concern. Neither would yield.
“I love you,” her mother said finally.
“I love you too, Mama. I’ll call on Friday.”
She returned to her patients. The woman with diabetes was stable. The child with pneumonia was responding to antibiotics. The motorcycle victim was in surgery. Ordinary emergencies, ordinary responses.
The hospital hummed with its usual rhythms. The systems functioned. The initial conditions held.
In London, Leila Mohammed finished her workday at a consulting firm in Canary Wharf.
She was twenty-eight, a senior data analyst, two years out of her Ph.D. Her work involved building predictive models for retail supply chains—algorithms that anticipated demand, optimized inventory, reduced waste. It was intellectually engaging and well-compensated and had nothing to do with Sudan, where she had not lived since she was eight years old.
Her parents had left in 1995, part of the brain drain that followed the coup that brought Omar al-Bashir to power. Her father was an engineer; her mother was a teacher. They had rebuilt their lives in London, raised their children in English schools, maintained their connections to Sudan through family calls and diaspora gatherings and a constant, low-grade homesickness that never fully resolved.
Leila’s connection to Sudan was mediated through screens. She followed the news. She sent money to relatives. She attended the occasional political event organized by expatriate groups. But her daily life was British—the Tube, the weather, the office politics, the Friday drinks with colleagues who knew nothing about Khartoum or Darfur or the intricate tribal politics that her parents still discussed at dinner.
She checked her phone on the commute home. A message from her cousin Mahmoud, who still lived in Khartoum:
Things are tense here. Pray for us.
She typed a response: What’s happening?
Negotiations collapsing. RSF moving troops. People are nervous.
She knew she should feel more. She knew that if something happened, her relatives would be in danger—the aunts and uncles and cousins who populated her parents’ stories, who appeared at weddings and funerals, whose faces she recognized from photographs but whose daily lives were abstract to her.
But she was tired. The models at work had bugs. Her flatmate was annoying. She had a date on Saturday that she was nervous about. Sudan was important, but Sudan was far away, and her capacity for sustained worry about distant crises had limits.
Stay safe, she typed. Let me know if you need anything.
She put the phone away. The train carried her toward home.
In a camp outside Khartoum, Bashir Hassan cleaned his rifle.
The ritual was meditative. Disassemble, inspect, oil, reassemble. The Kalashnikov was reliable machinery, robust against dust and neglect, but it performed better when maintained. Bashir had learned this in the Janjaweed. He had learned many things in the Janjaweed.
Around him, other RSF fighters went about their evening routines. Cooking fires. Card games. The low murmur of conversation. The camp was one of several staging areas near the capital, positioned according to a logic that the commanders understood and the soldiers did not question.
Bashir understood that things were about to change. The rumors flowed through the camp like water: the negotiations had failed, the SAF was planning to move against them, the commanders had decided to act first. Rumors were unreliable, but they were also the only intelligence available to common soldiers. You learned to read the patterns.
The commanders had been meeting more frequently. The supply deliveries had increased. The communications equipment had been tested repeatedly. These were not the behaviors of an organization expecting peace.
“Tomorrow,” a soldier named Youssef said, settling beside Bashir. “That’s what I heard.”
“You heard wrong last week.”
“This time is different. The General himself gave the orders.”
Bashir did not ask which general. The RSF had several. But if it came from the top—from Mohamed Hamdan Dagalo himself—then it was serious.
He finished cleaning his rifle. He checked his ammunition. He confirmed that his gear was ready. These were the preparations of a professional, executed without fanfare, without drama. If orders came, he would be ready.
He thought about his wife, about his children. He had sent money last month. He would send more if he survived what was coming. That was the arrangement: he did the work, he took the risks, and his family benefited from the RSF’s resources. It was not a bad life, as lives went. It was certainly better than herding cattle in the desert, which was where he had started.
The stars emerged over the camp. The desert night was cold. Bashir wrapped himself in a blanket and slept the sleep of a soldier who expected to wake to war.
Across Sudan, the initial conditions held.
Teachers taught. Doctors treated. Soldiers prepared. The negotiations continued in their final, futile hours. The forces were positioned. The intentions were fixed. The paths were chosen.
In the morning, the first transition would begin.
DEFINITION: A transition occurs when a system moves from one state to another according to its governing rules. The first transition is singular: it transforms initial conditions into dynamic evolution. After the first transition, there is no return to the prior configuration. The automaton has begun.
April 15, 2023 — 05:47 AM
The first explosion was not loud.
Maryam would remember this later, when people asked about the beginning. They expected drama—a thunderclap that announced the end of the world. But the sound that woke her was distant, muffled, almost ordinary. She had heard louder noises from construction sites, from the fireworks that marked Independence Day.
She lay in her bed for a moment, orienting herself. The dawn light was beginning to enter through the curtains. Saturday. Her day off. She had planned to visit her mother.
Then the second explosion came, closer, and she understood that the world had changed.
She was out of bed and dressed in ninety seconds. Her phone showed seventeen new messages, all arriving in the last three minutes. The resistance committee group: Shooting at the airport. SAF and RSF engaged. The hospital staff group: All personnel report immediately. Her mother: Maryam are you safe? What is happening? Call me.
She called her mother while walking to her car.
“I’m fine, Mama. I’m going to the hospital.”
“The radio says there’s fighting—”
“I know. I have to go.”
“Maryam, please—”
“I love you. I’ll call when I can.”
She drove through streets that were beginning to empty. People had the same information she did, were making the same calculations: stay home, protect family, wait for clarity. The drive that usually took twenty minutes took twelve. The checkpoints that should have been staffed were unmanned.
When she reached Ibrahim Malik Hospital, the first wounded were already arriving.
05:52 AM
Bashir had been awake for an hour when the orders came.
The camp had stirred before dawn. The soldiers knew something was happening; the commanders’ movements had signaled preparation. Bashir had eaten breakfast—flatbread and tea, the standard ration—and ensured his equipment was ready.
The commander gathered them at 05:30. A brief speech: the SAF had betrayed them, was planning to destroy the RSF, but they would strike first. Liberation. Victory. The usual vocabulary. Bashir had heard variations of this speech before campaigns in 2003, in 2013, in the years between. The words mattered less than the orders that followed.
His unit was assigned to the airport sector. They would link up with forces already positioned there. The objective was to seize the airport and deny it to SAF aircraft. Speed was essential.
They loaded into technicals—Toyota Land Cruisers mounted with heavy machine guns—and drove toward the city. The streets were nearly empty at this hour. The element of surprise was holding.
At 05:47, the first shots were fired at the airport. Bashir heard them from two kilometers away. By 05:52, his unit had engaged.
The SAF positions were stronger than expected. Soldiers in defensive positions, sandbags, crew-served weapons. The first technical in Bashir’s column took fire and swerved, its driver hit. Bashir’s vehicle pulled into cover behind a warehouse wall.
He exited and took position. The operation had begun.
06:15 AM
Ibrahim heard the news on the radio.
He had been preparing for morning prayers when the broadcast interrupted the usual recitation of Quran. A government announcement: military operations were underway in Khartoum. Citizens should remain calm. The situation was under control.
He looked at Khadija. Her face was still.
“It’s in Khartoum,” he said. “We’re far away.”
“It will come here.”
“You don’t know that.”
“I know.”
Her certainty was not prophecy. It was pattern recognition. Twenty years ago, she had watched violence spread from Khartoum’s political decisions to Darfur’s villages. The mechanism was reliable: when the center fractured, the periphery burned.
Ibrahim turned up the radio. The announcements were vague, reassuring in the way that lies are reassuring. He tried his mobile phone—calls to Khartoum went through intermittently, texts delivered with delays. His brother, who lived in Omdurman, responded after an hour:
Shooting everywhere. Staying inside. Will contact when safe.
The school had sent a message: closed until further notice. His students would not take their examinations. The future they had been preparing for had become suddenly uncertain.
“We should pack,” Khadija said. “Essential things only. If we need to leave quickly.”
He did not argue. He watched her move efficiently through the house, selecting items, organizing bags. She had done this before. Her body remembered even when her mind resisted remembering.
09:30 AM
Leila woke to chaos.
Her phone had accumulated overnight: family WhatsApp groups exploding with messages, news alerts from BBC and Al Jazeera, her mother’s frantic texts. She sat up in bed and scrolled, trying to piece together what had happened while she slept.
War. The word kept appearing. The RSF and SAF were fighting in Khartoum. The airport was a battleground. Explosions in the city center. Casualties unknown but mounting.
She called her mother, who was already awake and had been for hours.
“We’ve been trying to reach Mahmoud. The phones aren’t working properly.”
“What about Aunt Fatima? Uncle Hassan?”
“Nothing. We can’t get through.”
Leila felt the familiar distance collapse. Yesterday, Sudan had been an abstraction—news stories, family anecdotes, a place that existed in memory. Today, people she loved might be dying, and she was eight thousand kilometers away, unable to help, unable even to confirm that they were alive.
“What can I do?” she asked, and the question felt hollow.
“Pray,” her mother said. “And if you can send money—Mahmoud will need money if he has to leave.”
Money. The one form of assistance she could provide across distance. She opened her banking app and began researching transfer options. The usual services were unreliable during crises; she would need alternatives.
She spent the morning on the phone with relatives, with diaspora contacts, with anyone who might have information. The picture that emerged was fragmentary but consistent: heavy fighting across Khartoum, both sides taking casualties, civilians caught between. No one knew how long it would last. No one knew what came next.
14:00 PM
By afternoon, the hospital was overwhelmed.
Maryam had worked through the morning without pause. The wounded arrived in waves—soldiers at first, then civilians caught in crossfire, then more civilians as the fighting spread through residential neighborhoods. The triage system she had learned in training became operational necessity: red for immediate, yellow for urgent, green for minor, black for dead.
Too many reds. Not enough surgeons. Not enough blood. Not enough anything.
She had assisted with three amputations before lunch. The surgical team worked with what they had: anesthetic running low, antibiotics rationed, sutures counted and recounted. The power failed twice; the generator kicked in both times. When the generator fuel ran out, they would be operating by flashlight.
A boy came in around two o’clock. He was perhaps twelve, with a fragment wound to his abdomen. His mother was with him, silent with shock, unable to answer the intake questions. Maryam took the boy to surgery and watched the doctors open him and try to repair what could not be repaired. He died on the table at 14:47.
She wrote the time on his chart. She covered his body. She told his mother.
Then she returned to the next patient, and the next, and the next.
This was the first day. The automaton had begun, and she was inside it.
17:00 PM
The airport had not fallen.
Bashir’s unit had fought for eleven hours. They had advanced, retreated, advanced again. The SAF had reinforced their positions; the RSF had brought more technicals. The casualties were heavy on both sides.
By evening, the front line had stabilized. The RSF controlled the perimeter road. The SAF held the airport facilities. Neither side could push further without unacceptable losses.
Bashir rested against a concrete barrier, drinking water from a plastic bottle. His hands were steady. They were always steady after combat—a physiological quirk that he had learned to rely on. The shaking came later, in private, when the body processed what the mind had refused.
A soldier he didn’t know sat beside him. Young, maybe twenty. The young ones talked after their first engagements; they needed to process.
“I didn’t think it would be like this,” the soldier said.
Bashir didn’t respond. What was there to say? Combat was what it was. The surprise wore off; the familiarity set in. The operations became routine.
“What happens now?” the soldier asked.
“We hold. We wait for orders. We take what they tell us to take.”
“And if they tell us to take the city?”
“Then we take the city.”
The young soldier fell silent. Perhaps he was imagining what taking the city would mean. Perhaps he was not ready to imagine it. Bashir had been his age once, had asked similar questions, had received similar answers. The understanding came with experience.
Night fell over Khartoum. The gunfire continued, sporadic, distant. The first day was ending, but the automaton would continue. The transitions would accumulate. The states would change.
23:00 PM
Yusuf received orders in Port Sudan.
He had spent the day in the headquarters, monitoring reports, participating in coordination calls. The situation in Khartoum was dire. The RSF offensive had been larger and more coordinated than anticipated. Key facilities were contested. The chain of command was strained.
His transfer to El Fasher was accelerated. He would depart tomorrow rather than in two weeks. The garrison there needed reinforcement, and he was the available senior officer with relevant experience.
Ismail briefed him on the route. The direct roads might be compromised; they would fly if aircraft were available, drive if not. El Fasher was 900 kilometers away, and 900 kilometers of uncertainty lay between.
“What’s the situation there?” Yusuf asked.
“Calm for now. But if the RSF has launched a coordinated offensive, Darfur will follow. They have forces positioned throughout the west.”
“Then we’ll be defending.”
“Yes, sir. The question is for how long.”
Yusuf looked at the map. El Fasher was isolated. The supply lines were long and vulnerable. If the RSF controlled the roads, the garrison would depend on air resupply—and air resupply depended on holding the airport, which was precisely what was failing in Khartoum.
He understood the mathematics. A siege was possible. A siege was likely. And sieges were wars of attrition, determined not by courage but by resources.
“Pack for a year,” he told Ismail.
“A year, sir?”
“At least.”
At midnight, the first day ended.
Across Sudan, the initial conditions had given way to dynamic evolution. The automaton was running. The states were changing. The primitives were composing into procedures.
In Khartoum, the dead lay where they had fallen, and the living prepared for morning.
In El Geneina, Ibrahim and Khadija slept uneasily, their bags packed by the door.
In London, Leila stared at her phone, waiting for messages that did not come.
In the hospital, Maryam worked through the night, counting the living and documenting the dead.
In the camp outside the airport, Bashir cleaned his rifle and waited for orders.
The first transition was complete. The system was in motion.
There would be no return to April 14.
Definition: A primitive carrier is a human substrate capable of executing fundamental operations. The carrier transforms abstract procedure into physical action. The same individual may carry different primitives at different times; the same primitive may be carried by many individuals. The carrier is not the primitive—but without carriers, primitives cannot propagate.
Bashir Hassan was not born a killer.
He was born in 1978 in a village in North Darfur, the third son of a livestock trader who moved with the seasons. His earliest memories were of dust and cattle, the rhythm of migration that had defined his people for centuries. The rains came; they moved south. The rains ended; they moved north. The pattern was older than nations.
He learned to ride before he could read. He learned to shoot at twelve—the rifle was his father’s, a bolt-action thing that had passed through three generations. Marksmanship was practical: the predators that threatened livestock did not respect treaties or borders. A boy who could hit a moving target at 200 meters was valuable. Bashir was valuable.
The first time he killed a man, he was seventeen.
It was during the conflicts over grazing rights that periodically flared between Arab herders and African farmers. The disputes were ancient—older than the British, older than the Mahdiyya, older than the Funj Sultanate. But the droughts of the 1980s had compressed everyone onto smaller and smaller lands, and the ancient disputes had become lethal.
The man he killed was Fur. A farmer who had accused Bashir’s uncle of allowing cattle to graze on cultivated land. There had been an argument, then threats, then weapons. Bashir’s bullet struck the man in the chest from perhaps fifty meters. The man fell and did not rise.
What Bashir remembered, afterward, was not guilt. It was the precision of the moment—the alignment of eye and sight and target, the squeeze of the trigger, the mechanical certainty of cause and effect. The man had been a problem. The bullet had been a solution. The problem was solved.
This clarity would shape his subsequent life.
The Janjaweed found him in 2003.
By then, the government in Khartoum had decided that the insurgencies in Darfur required a different approach. The army was stretched thin; conventional operations were expensive and slow. But there were men in Darfur who already knew how to fight, already knew the terrain, already had grievances against the farmers and villagers who supported the rebels.
A recruitment officer came to Bashir’s camp. The offer was straightforward: weapons, supplies, payment, and license. The targets would be rebel supporters. The methods would be flexible.
Bashir understood what “flexible” meant. He had seen the burned villages, the bodies by the wells. He had heard the stories of what the horsemen did when they arrived. The operations were not military in any conventional sense. They were punitive. They were collective.
He joined anyway.
Why? He had asked himself this question many times over the subsequent years. The answers shifted depending on when he asked. Sometimes it was economics—the payment was better than herding. Sometimes it was tribal loyalty—his people were at war, and one fought for one’s people. Sometimes it was simpler: the government had offered him a gun and permission to use it, and he had discovered at seventeen that using guns produced clarity.
Perhaps all the answers were true. Perhaps none of them were. The Janjaweed did not require motivation; they required compliance. Bashir was compliant.
The operations had a structure.
First came the aircraft. Government Antonovs would bomb a village, sometimes hitting, sometimes missing, always terrifying. The bombs announced that the state had marked this place for destruction.
Then came the horsemen.
Bashir rode in at dawn. His unit approached from the east, silhouetted against the rising sun so the defenders could see them coming. There was psychological purpose to this: the fear was itself a weapon.
The procedure followed a template:
Surround the village. Cut off escape routes. Anyone who fled would be pursued and killed, but fewer would flee if they believed escape was impossible.
Identify. The men were separated from the women and children. Anyone with weapons was killed immediately. Anyone who resisted was killed immediately.
Interrogate. Where were the rebels? Who had supported them? Who had hidden them? The answers mattered less than the questions. The questions established that cooperation was expected; the punishment for non-cooperation was already determined.
Execute. The men—all of them, or most of them, or enough of them to ensure that the village would never again produce fighters. The executions were public. The women and children watched.
Loot. Cattle, grain, whatever had value. The livestock was driven away. The grain stores were destroyed if not taken.
Burn. Everything flammable. The houses, the crops, the bodies sometimes. The smoke rose and could be seen from the next village, which would understand what was coming.
Bashir performed his function in this machinery. He did not design the operations; he executed them. He was good at execution. His shooting remained precise. His hands remained steady. The clarity he had discovered at seventeen extended to encompass larger and larger operations.
By 2005, he had participated in the destruction of forty-seven villages. He had personally killed an uncounted number of men—uncounted because counting was not required, was perhaps impossible given the chaos of the operations. The bodies accumulated behind him, and he moved forward.
This is what it means to carry a primitive. The carrier does not invent the operation; the carrier implements it. The carrier is the hardware on which the software runs.
In a hospital supply closet in Omdurman, in April 2023, a woman named Aisha Yusuf was becoming a different kind of carrier.
She was twenty-six years old, a pharmacy technician who had worked at this hospital for four years. Her job was inventory: counting pills, organizing supplies, ensuring that the medical machinery had the inputs it required. She was good at systems.
When the war began, the hospital transformed. The ordinary patients—the scheduled surgeries, the chronic conditions, the routine births—were suddenly displaced by trauma. Gunshot wounds. Shrapnel injuries. Blast damage. The pharmacy that had been ordering blood pressure medications was now rationing morphine.
Aisha discovered that her skills transferred. Inventory management was inventory management, whether the inputs were antibiotics or bandages. The system was overwhelmed, but systems could be optimized. She began tracking what the hospital had, what it used each day, what it needed. She created a spreadsheet, then a database, then a network of contacts who could source supplies when the official channels failed.
Within a week, she was coordinating with other hospitals. Within two weeks, she had connections to emergency response rooms in three neighborhoods. The ERR volunteers needed supplies; the hospitals had some supplies but lacked distribution capacity. Aisha became the junction point, the router through which information and material flowed.
She was carrying the counter-primitives: identify need, coordinate response, deliver assistance, adapt to disruption, document everything.
She had not chosen this role consciously. The role had emerged from the application of her existing skills to new circumstances. She was still doing inventory management; the inventory had merely expanded to encompass the survival of her neighbors.
This is also what it means to carry a primitive. The carrier need not understand the larger pattern. The carrier need only execute the operation in their local context. The primitive spreads because carriers emerge wherever the conditions favor their emergence.
The RSF that absorbed Bashir in 2013 was different from the Janjaweed he had joined in 2003.
The name had changed—Rapid Support Forces rather than mounted militias. The formal status had changed—integrated into the state security apparatus rather than deniable auxiliaries. The equipment had changed—technicals with mounted guns rather than horses, though horses were still used for certain operations in certain terrain.
But the primitives remained constant.
The new organization added layers of bureaucracy. There were ranks now, salary scales, official deployments. Bashir rose to command a unit of thirty men. He received orders through a chain of command that extended to Khartoum. The operations were documented, at least partially; there were reports to file, requisitions to submit.
The core remained unchanged: identify, separate, execute, loot, erase. The villages were different—sometimes Nuba communities rather than Fur or Masalit, sometimes alleged rebel sympathizers rather than confirmed insurgents—but the procedure was the same procedure he had learned twenty years before.
Bashir understood that he had become an institutional carrier. His knowledge, his experience, his proven reliability—these were assets that the organization cultivated. When younger soldiers joined, he trained them. He transmitted the primitives to new substrates. The pattern replicated.
The month after the war began, Leila Mohammed attended a meeting in East London.
The gathering was organized by Sudanese diaspora activists—people who had left Sudan years or decades ago but had maintained connections. The topic was aid coordination. The need in Sudan was vast; the diaspora response was fragmented. How could they help more effectively?
Leila had come because her mother asked her to. She was not an activist; she was a data scientist who worked on fraud detection for a financial services company. Her expertise was pattern recognition in transaction data. She did not see how this translated to humanitarian coordination.
But she listened, and as she listened, she recognized something familiar.
The speakers described the challenges: multiple organizations working without coordination, resources duplicated in some areas and absent in others, information scattered across WhatsApp groups and spreadsheets and individual memories. They needed systematic tracking. They needed data integration. They needed pattern recognition to identify where need was greatest and response most lacking.
Leila raised her hand. “I might be able to help with that.”
Within three weeks, she had built the first version of what would become the diaspora coordination platform. It aggregated reports from ERR groups across Sudan, tracked donations and distributions, identified gaps in coverage. The database grew as volunteers added information; the analysis improved as patterns emerged.
She was still doing her day job, still detecting fraud in financial transactions. But in her evenings and weekends, she was detecting something else: the distribution of suffering, the flow of aid, the points where the system broke down.
She had become a carrier without leaving London. The counter-primitives could propagate through data as well as through direct action.
Not all carriers were willing.
In Khartoum, a young man named Tariq Hassan was conscripted into the RSF after the war began. He was nineteen, a university student who had been studying engineering before the campus closed. He had no military experience, no tribal connection to the RSF’s power base. He was simply present and male and of military age.
The RSF was expanding rapidly. The initial forces—the Janjaweed veterans, the professional soldiers, the true believers—were insufficient for the scale of operations required. They needed bodies. Tariq was a body.
He was given a rifle and minimal training. The training was not about marksmanship; it was about obedience. Follow orders. Stay with your unit. Do what you’re told. The primitives would be transmitted through practice.
His first operation was a house search in a neighborhood suspected of harboring SAF sympathizers. The procedure was familiar to the veterans—familiar because it was the same procedure used in Darfur, adapted for urban terrain. Enter the house. Separate the men. Search for weapons or evidence of military connection. Take what was valuable. Leave.
Tariq’s hands shook as he pointed his rifle at a family huddled in their living room. A father, a mother, two children. They looked like his neighbors. They probably were his neighbors—Khartoum was not that large, and everyone knew someone who knew someone.
He followed orders. He did not shoot anyone that day. But he watched others shoot, watched the primitive execute through substrates who did not hesitate. He understood that hesitation was discouraged. He understood what happened to soldiers who refused orders.
He was becoming a carrier against his will, the primitive forcing itself into him through pressure and example and survival instinct. The system did not require enthusiasm. It required compliance.
The carriers shaped the primitives even as the primitives shaped the carriers.
Bashir, in twenty years of operations, had developed refinements. He knew which methods produced results and which produced only screaming. He knew how to manage a scene to minimize resistance. He knew that certain kinds of violence paralyzed and certain kinds provoked. This knowledge accumulated in him and transmitted to those he trained.
Aisha, in two months of coordination, had developed her own refinements. She knew which supply chains were reliable and which promised more than they could deliver. She knew which ERR groups communicated well and which disappeared for days. She knew the geography of Khartoum’s needs as precisely as the military knew its geography of targets.
The carriers were not interchangeable parts. Each one brought individual capacity, individual limitations, individual adaptations. The primitives expressed differently through different substrates. But the core operations—the fundamental procedures—remained recognizable.
This is the nature of execution on human hardware. The software runs, but the hardware matters. The same algorithm produces different outputs depending on the system it runs on. And yet the algorithm remains the algorithm.
In El Geneina, as April turned to May, Ibrahim watched the carriers arrive.
They came in waves. First the RSF regulars—organized units with command structure and logistical support. Then the tribal militias—less organized, more enthusiastic, drawn by promises of loot and ethnic grievance. Then the individuals—young men with guns who attached themselves to whatever unit would have them.
He recognized what he was seeing. Khadija had described this to him, in the fragments she could bear to speak. The buildup before the storm. The accumulation of carriers at the edge of the target zone.
“They’re not here for the rebels,” she said. “There are no rebels here. They’re here for us.”
Ibrahim wanted to argue. The RSF had made statements about counterinsurgency operations, about restoring order, about protecting civilians. The governor had negotiated, had sought assurances, had tried to establish channels of communication.
But he saw the young men with guns, saw how they looked at Masalit neighborhoods, saw the calculations in their eyes. They were carrying something, and what they carried would express itself soon.
“We should leave,” Khadija said.
“Where would we go?”
“Chad. We have family in Adré. We know the way.”
The border was seventy kilometers away. In normal times, a journey of two hours. But the roads were uncertain now. And leaving meant abandoning everything—the house they had built, the school where he taught, the community that had welcomed them twenty years ago.
“Give it one more week,” he said. “The situation might stabilize.”
Khadija did not argue. She simply looked at him with the eyes of someone who had seen this before, who knew that stabilization was not what was coming.
The carriers were already in position. The primitives were already loaded. The only question was when the system would begin execution.
A primitive without a carrier is an abstraction. A carrier without a primitive is merely a person. The fusion produces capacity—the ability to perform operations at scale, to implement procedures across populations, to transform individual actions into systemic effects.
The war in Sudan was building an army of carriers.
On one side, the RSF was training, recruiting, conscripting—expanding the substrate that could execute its procedures. Veterans like Bashir transmitted knowledge to newcomers. The primitives replicated through instruction and practice and the raw pressure of operational necessity.
On the other side, the ERR networks were growing their own carrier population. Volunteers like Aisha developed skills and transmitted them to others. The counter-primitives propagated through example and training and the same pressure of necessity.
Both systems were scaling. Both were recruiting human beings and transforming them into executors of procedure. The outcome of the war would depend, in part, on which system could build carriers faster, deploy them more effectively, sustain their operation longer.
But there was an asymmetry that no recruitment could erase.
The genocide primitives required carriers who would kill their neighbors. This was a high barrier. Many people could not become such carriers regardless of pressure or incentive. The pool of potential genocide carriers was limited by something fundamental in human psychology.
The counter-primitives required carriers who would help their neighbors. This was a much lower barrier. Most people, given opportunity and minimal support, would help. The pool of potential counter-carriers was nearly unlimited.
The RSF could never recruit as many killers as the ERRs could recruit helpers. The mathematics favored resistance.
But mathematics alone did not determine outcomes. Killers with guns were more powerful, in the immediate term, than helpers with supplies. The counter-carriers could be killed by the genocide carriers. The asymmetry of capacity did not translate automatically to asymmetry of victory.
The system would run, and the carriers would carry, and the interaction of carriers and primitives would produce whatever it produced.
At midnight on May 15, 2023, exactly one month after the war began, the carriers were distributed across Sudan like charges in a circuit.
Bashir was in Darfur, attached to a unit moving toward El Geneina. His role would be the role he had always played: the reliable executor, the carrier of procedure.
Aisha was in Omdurman, her supply network now spanning seven hospitals and forty-three neighborhoods. She had not slept more than four hours a night in three weeks. The need was infinite; her capacity was finite.
Leila was in London, her database tracking 200,000 displaced persons and their needs. She had learned to read the patterns in the data—to see where violence was concentrating, where famine was building, where the next crisis would emerge before it emerged.
Tariq was in Khartoum, still holding a rifle he didn’t want, still following orders he didn’t believe in, still becoming something he had never intended to become.
Ibrahim and Khadija were still in El Geneina, still waiting, still hoping for the stabilization that would not come.
Colonel Yusuf was in El Fasher, fortifying positions, counting supplies, calculating how long his garrison could hold against siege. Ismail was beside him, doing the same calculations, reaching the same troubling conclusions.
Maryam was in the hospital, still working, still counting the dead and the living, still carrying the primitive of documentation even when there was no one to document for.
The carriers were in position. The primitives were loaded. The automaton was processing.
June was coming.
Definition: A primitive resource is an input that enables the execution of other primitives. Without primitive resources, the system cannot operate. Gold is the primitive resource of the Sudan genocide—the substrate that converts earth into weapons, minerals into murder. Control of gold determines capacity for sustained operation.
Jebel Amer rises from the desert of North Darfur like a broken promise.
The mountain is not impressive by geological standards—a few hundred meters of rock and sand, distinguishable from surrounding terrain only by its slight elevation. But beneath its surface lies one of the largest gold deposits in East Africa, and this geological accident has shaped the fate of millions.
Gold was discovered here in 2012. Within months, a hundred thousand miners had arrived. Within years, the number reached three hundred thousand—a city of tents and shacks and makeshift processing facilities, appearing in the desert like a tumor.
The Janjaweed—not yet the RSF, still the irregular militia of the Darfur genocide—recognized the opportunity immediately. A mountain of gold, accessible, ungoverned. An economic base that could transform a militia into an army, a temporary formation into a permanent institution.
Hemedti understood this better than anyone.
Mohamed Hamdan Dagalo, known as Hemedti, had risen through the Janjaweed by a combination of ruthlessness and calculation. By 2012, he commanded thousands of fighters. But fighters required payment, and payment required revenue, and the government in Khartoum was an unreliable patron. The gold offered independence.
His forces moved on Jebel Amer systematically. The tribal conflicts over mining rights became opportunities for intervention. The RSF positioned itself as a stabilizing force, bringing order to chaos—and extracting tribute for that order. By 2017, the mountain was effectively under RSF control.
The annual production reached an estimated 80 to 100 tons. At global prices, this represented $4 to $5 billion in raw value. Not all of it flowed to the RSF—the miners took their share, the traders took their share, the government in Khartoum demanded its cut. But even a fraction of billions is wealth sufficient to build an army.
This is the gold primitive: earth converted to power through the mediation of global markets.
Dubai is eight thousand kilometers from Jebel Amer, but the distance is measured in hours, not geography.
The gold moves in chartered flights, in overland convoys, in the false bottoms of ordinary vehicles. It crosses borders with documentation that is accurate enough to pass inspection and vague enough to obscure origins. It arrives in the UAE bearing certificates of legal provenance, and the UAE does not inquire too closely into the definition of “legal.”
Leila discovered this supply chain three months into the war, when her tracking of RSF capabilities revealed impossible logistics.
The RSF was importing weapons, vehicles, fuel, ammunition—far more than the official Sudanese military budget could explain. They were paying salaries to fighters, maintaining supply lines, sustaining operations across thousands of kilometers. Where was the money coming from?
She followed the data. Export records showed Sudanese gold flowing to Dubai in quantities that far exceeded legal mining production. Import records showed the UAE as Sudan’s primary trading partner for gold, absorbing over 90% of declared exports. But the declared exports were a fraction of actual production.
The gap was the RSF’s operating margin.
“They’re self-financing,” Leila told the diaspora coordination meeting. “They don’t need outside support to continue fighting. They have gold.”
The implications were stark. A militia that depended on government funding could be constrained by cutting that funding. A militia that generated its own revenue through resource extraction was independent—economically autonomous, capable of sustained operation regardless of political shifts.
“The sanctions won’t work,” she said. “Not unless they target the gold trade. And the gold trade runs through Dubai.”
The room was silent. Dubai was a major global financial center. The UAE was a strategic ally of Western powers. The gold trade was legal, technically—the UAE had no obligation to verify the origins of every shipment.
“They’ve built a machine,” Leila continued. “Gold comes out of the ground. Gold goes to Dubai. Dubai converts gold to dollars. Dollars buy weapons. Weapons enable extraction of more gold. The loop is closed.”
This is the power of a primitive resource: it creates self-sustaining cycles. The system does not require external input to continue operating. The system feeds on itself.
In the mines of Jebel Amer, a man named Osman Idris was digging.
He was fifty-two years old, a farmer from South Darfur whose land had been destroyed in the 2003 genocide. He had survived—barely—and wandered for years through camps and settlements before finding his way to the gold rush. Mining was brutal work, but it paid better than anything else available to a displaced person with no land and no credentials.
The primitive operation of extraction was simple. You dug. You crushed. You washed. The gold separated from the ore by virtue of its density, collecting in the bottom of the pan while lighter material washed away. You collected what you found and sold it to the traders who circulated through the camps.
The traders paid in cash, which was useful. But the prices they offered were far below market value, and there was no alternative. The RSF controlled access to the mountain. The RSF controlled the roads out. The RSF took a cut of every transaction. You sold at the price offered or you didn’t sell.
Osman did not know where his gold went. He did not know about Dubai, about the refineries, about the weapons purchased with the proceeds of his labor. He knew only that he dug and crushed and washed, and that at the end of the week he had enough money to buy food and perhaps send something to his daughter in Chad.
He was a node in the supply chain, a carrier of the extraction primitive, but he did not understand the system of which he was part.
This is also characteristic of primitive systems: the operators need not comprehend the operation. They need only execute their local function. The understanding exists in the structure, not in the nodes.
The gold primitive enabled the other primitives.
Consider the economics of genocide. Killing is expensive. You need weapons, ammunition, vehicles, fuel. You need to pay fighters—even ideologically motivated fighters require food and shelter and some incentive beyond pure hatred. You need logistics to move forces across large distances. You need communication to coordinate operations.
The Rwandan genocide of 1994 killed 800,000 people in 100 days, primarily with machetes. The low-technology approach was partly circumstantial—Rwanda was poor, and machetes were available—but it was also a limitation. The killing was inefficient in military terms. Significant portions of the target population survived simply by fleeing.
The Sudan genocide operated at industrial scale because it was funded at industrial scale. The RSF could afford drones, artillery, technical vehicles with mounted weapons. They could move rapidly across the theater, pursuing populations that tried to flee. They could sustain sieges for months or years, maintaining the logistical infrastructure necessary for prolonged operations.
Gold bought this capacity.
In El Fasher, Colonel Yusuf understood the mathematics. His garrison had limited supplies, limited ammunition, limited fuel. The RSF besieging him had functionally unlimited resources—or at least resources sufficient to sustain pressure indefinitely. The siege was a war of attrition, and attrition favored the side with the larger economic base.
“We’re not fighting an army,” he told Ismail in May 2024, a year into the siege. “We’re fighting a mining company.”
The analysis was not quite right—the RSF was more than a mining company, carried ideological and ethnic dimensions that pure extraction could not explain. But the economic core was accurate. The gold made everything else possible.
The counter-economy operated on different principles.
The ERRs did not have gold. They had diaspora remittances, community contributions, and the occasional grant from international organizations. Their funding was fragmented, inconsistent, dependent on the goodwill of thousands of individual donors.
But they had discovered something the RSF had not: the economics of solidarity were more efficient than the economics of extraction.
Consider the cost structures. The RSF paid fighters, purchased weapons, maintained vehicles. Their operating costs were high because their operations required specialized equipment and trained personnel. The gold financed these costs, but the costs were substantial.
The ERRs paid nothing to their volunteers. Their equipment was improvised—WhatsApp groups instead of military communications, donkeys instead of technicals, community kitchens instead of supply chains. Their operating costs were minimal because their operations relied on freely given labor and repurposed civilian infrastructure.
Aisha calculated this in June 2023, during a brief period of reflection between supply runs. The RSF spent thousands of dollars per operation. The ERRs spent hundreds. The RSF needed continuous gold revenue to sustain itself. The ERRs needed only continuous willingness to help.
“They have more money,” she told her network. “But we have more people. And people don’t run out.”
This was the asymmetry at the economic level: extraction versus contribution, scarcity versus abundance. The gold primitive was powerful but finite. The solidarity primitive was weaker per unit but potentially infinite.
The hawala system connected the two economies.
Hawala is an ancient mechanism for transferring value across distance without moving physical currency. A person deposits money with a broker in one location; the broker contacts a counterpart in another location; the counterpart pays out the equivalent amount. No money crosses borders. Only trust and obligation move.
The diaspora used hawala to send funds to Sudan. Leila’s coordination platform tracked these flows—she could see patterns in the data that revealed which networks were reliable, which routes were secure, which communities were receiving support.
The RSF used hawala as well. Some of their gold revenue flowed through the same networks, converted to dollars in Dubai and transferred to operatives across the region. The two economies—genocide and resistance—operated through the same financial infrastructure.
“We’re using the same pipes,” Leila observed. “But pushing different fluids.”
The observation was unsettling but structurally accurate. The informal finance system was neutral. It served whoever paid for its services. The gold primitive and the solidarity primitive both required mechanisms for value transfer, and both used whatever mechanisms were available.
This neutrality was characteristic of primitive resources. Gold did not care who dug it up or what it was used for. The hawala network did not distinguish between transactions funding murder and transactions funding medicine. The primitives were amoral—their morality emerged only in application.
By June 2023, the gold had bought an army outside El Geneina.
Ibrahim watched them arrive from the window of his house, which overlooked the road from Khartoum. Convoys of technicals, each mounted with heavy weapons. Trucks loaded with supplies. Fuel tankers to sustain the vehicles. The logistics were professional, precise, indicative of funding that could afford precision.
“They’re preparing something large,” Khadija said. She had been watching too, her face drawn with memories of 2003.
“The governor is negotiating. There are channels—”
“There were channels before. The channels didn’t help.”
Ibrahim wanted to argue. The situation was different—the international community was aware, the documentation was ongoing, the world was watching. But the convoys kept coming, and each convoy represented gold converted to killing capacity.
He did not understand, then, the economic substrate of what he was witnessing. He did not know about Jebel Amer or Dubai or the closed loop of extraction and expenditure. He saw only the surface manifestation: weapons accumulating, fighters massing, the shadow of violence growing longer with each passing day.
“We should leave,” Khadija said again.
“Tomorrow. If the situation worsens.”
The situation worsened that night. The RSF established checkpoints on the roads out of the city. The escape routes closed, one by one, until the only way out was through hostile forces or across seventy kilometers of desert to Chad.
The gold had purchased encirclement. The siege had begun.
In the larger view—the view that Leila could sometimes glimpse through her data, that von Neumann would have recognized as the view of system dynamics—the gold primitive was the key constraint.
Remove the gold, and the RSF became an ordinary militia: dangerous but limited, capable of local violence but not sustained campaign. The transformation from Janjaweed to RSF, from irregular raiders to industrial-scale genocide operation, had been enabled by control of Jebel Amer.
But removing the gold was not simple.
The gold was physically present—tons of it extracted monthly, transported along established routes, processed through known channels. The physical removal would require military control of the mining region, which no one was positioned to achieve.
The gold trade was legally ambiguous. The UAE was not violating any explicit international law by purchasing Sudanese gold. The trade was distasteful but not prohibited. Sanctions would require international coordination that was not forthcoming.
The gold market was global. Even if one destination closed, others would open. Dubai was convenient but not unique. Gold was fungible, liquid, universally valued. The RSF could find buyers.
The primitive resource was embedded in structures larger than any single actor could control. The gold flowed because global economics demanded that it flow. The genocide was funded because the funding mechanisms existed and operated efficiently.
“This is what we’re up against,” Leila told her mother during one of their weekly calls. “Not just men with guns. A whole system that makes men with guns profitable.”
“Then fight the system.”
“How?”
Her mother did not have an answer. No one had an answer. The gold primitive was too deeply rooted, too widely connected, too structurally essential to the global economy of extraction and exchange.
The automaton continued processing. The gold continued flowing. The weapons continued arriving at the outskirts of El Geneina.
There is a mathematical property of certain systems called positive feedback: a process that amplifies itself, that uses its outputs as inputs to generate more outputs.
The gold primitive created positive feedback in the genocide system.
More gold → more weapons → more territory → more gold mines → more gold.
The cycle had no internal limiting factor. As long as the RSF could convert gold to military capacity and military capacity to territorial control and territorial control to access to more gold, the process would continue. The feedback was positive in the mathematical sense—not beneficial, merely amplifying.
The counter-system also had feedback, but of a different character.
More solidarity → more volunteers → more aid delivered → more trust → more solidarity.
This cycle was also self-amplifying, but it did not produce the same material capacity. More solidarity did not directly generate more weapons or more territorial control. The ERRs grew, but they grew in capability for aid delivery, not for military resistance.
The two feedback systems operated in parallel, each amplifying, each growing. The question was which would reach critical mass first—whether the genocide machine would complete its operation before the solidarity network could build sufficient infrastructure to survive it.
On June 9, 2023, the gold-funded operation against El Geneina began.
Ibrahim heard the first explosions at dawn. He had not slept—the accumulating evidence of what was coming had kept him awake, watching from the window, as if watching could prevent. The explosions were distant, perhaps at the airport, perhaps at the military garrison.
Khadija was already moving. Twenty years of trauma had not erased her competence in crisis; it had honed it. She gathered their bags, checked the water containers, ensured that the documentation Ibrahim had been collecting—testimonies, photographs, records—was secure.
“The western road,” she said. “If we move now, before they establish full perimeter.”
Ibrahim hesitated. The school. The students. The community that had relied on him for twenty years.
“They’re killing everyone,” Khadija said. “The teachers first. The ones who might remember, who might tell what happened. You know this.”
He did know it. The pattern was documented, analyzed, understood. The genocide targeted the articulators—the educated, the leaders, the witnesses. A teacher who recorded testimonies was a priority target.
“Give me one hour,” he said. “To gather what I can.”
“We don’t have an hour.”
She was right. They both knew she was right. But Ibrahim could not leave without trying to preserve something of what he had witnessed. The notebooks on his desk contained months of documentation. The testimony of survivors. The names of the dead.
He grabbed what he could carry. They left their house at 6:14 AM, moving toward the western edge of the city, toward Chad, toward uncertain safety.
Behind them, the gold-funded machinery began its work.
Definition: Nafeer (Arabic: نفير) is a call to collective action embedded in Sudanese culture. It is a primitive resource of a different kind than gold—not material but social, not extracted but invoked. Nafeer converts individual capacity into collective response through shared understanding. Where gold enables destruction, nafeer enables construction.
The tradition is older than Islam in Sudan, older than the kingdoms along the Nile, older than recorded history.
When the floods came and the crops were threatened, the village gathered to build dikes. When a family lost their home to fire, the neighbors built a new one. When drought struck and livestock died, those who had shared with those who had not. The word for this was nafeer, and the word contained both the call and the response—the request for help and the obligation to provide it.
Samira Abdalla learned nafeer from her grandmother.
She was eight years old, in a village outside Omdurman. A neighbor’s house had collapsed during the rains, and the family was living under a tarp while their belongings turned to mud. Samira’s grandmother took her by the hand and led her to where the men of the village were gathering.
“Watch,” her grandmother said.
What Samira watched was transformation. The men arrived with tools—some brought shovels, some brought ropes, some brought only their hands. There was no foreman, no plan distributed in advance, no payment offered or expected. The men assessed the damage, divided the labor, and began working. By evening, the frame of a new house was standing. By the next evening, the walls were up. By the third day, the family moved in.
“This is nafeer,” her grandmother said. “Someone asks. Everyone comes. The work gets done.”
“Why?”
“Because someday you will ask, and they will come for you.”
The reciprocity was not transactional in the market sense. There was no ledger tracking who had helped whom. The obligation was general: help your neighbor, and trust that your neighbors will help you. The accounting was social, not economic.
Samira internalized this at eight years old. It became part of her operating assumptions, a primitive loaded into her cognitive substrate before she had language to describe it.
The 2019 revolution ran on nafeer.
When the protests began against Omar al-Bashir—the dictator who had ruled Sudan for thirty years, who had sheltered Osama bin Laden, who had presided over the Darfur genocide—the Resistance Committees formed. These were neighborhood organizations, spontaneous and decentralized, emerging wherever people gathered to coordinate.
Samira was twenty-seven then, working as an accountant for a shipping company, bored and angry and ready for something different. She joined the Resistance Committee in her neighborhood of Bahri, north of the river, and discovered that the organizational skills she used for shipping logistics translated directly to protest logistics.
Where would the marchers assemble? What routes avoided police checkpoints? Who had water to distribute? Who could provide first aid if the security forces fired tear gas? The questions were practical, and practical questions had answers, and answers came from collective problem-solving.
The protests succeeded—partly. Bashir fell. But the military remained in power, negotiating a transition that always seemed to recede. The Resistance Committees persisted, maintaining their networks, waiting for the next phase.
When the war began, the networks activated.
Samira’s WhatsApp group—originally used for protest coordination—became a humanitarian hotline within the first day. People posted their needs: trapped in their apartments, running out of food, needing evacuation. Other people responded: I have a car, I have supplies, I know a safe route. The matching happened organically, without central direction.
“This is nafeer,” Samira told a journalist who interviewed her six months later. “We have been practicing this our whole lives. The war just gave us new problems to solve.”
Nafeer as primitive resource had properties opposite to gold.
Gold was scarce. Nafeer was abundant—potentially unlimited, bounded only by the willingness of people to participate.
Gold could be hoarded. Nafeer could not be stored; it existed only in the act of response. A community that did not practice nafeer could not retrieve it when needed.
Gold was fungible. Nafeer was contextual—the specific response matched to the specific need, adapted to local conditions and capabilities.
Gold required extraction infrastructure. Nafeer required only communication—a call, a message, a signal that help was needed.
The ERR networks understood these properties intuitively, even if they could not articulate them formally. They built their operations around the nafeer primitive because that was the resource available to them. They did not have gold. They had each other.
Leila, in London, struggled to understand nafeer from the outside.
Her analytical training demanded quantification. How many ERR volunteers were there? What was their capacity? How did coordination work? She built dashboards and databases, trying to capture the structure of what she was observing.
But the structure resisted capture. The ERR networks were not hierarchical organizations with clear roles and responsibilities. They were fluid, self-organizing, responsive. A person who coordinated evacuations one week might run a community kitchen the next. A neighborhood that received aid one month might become an aid hub the following month.
“It’s not an organization,” she complained to her mother. “It’s more like... weather. A weather system that decides where to send help.”
“It’s nafeer,” her mother said.
“What does that mean, exactly? Operationally?”
Her mother was quiet for a moment. “When I was a girl, and the harvest was good, my father would give grain to families whose harvest was poor. No one told him to. There was no program. He did it because that was what you did.”
“And if your harvest was poor?”
“Others would give to us.”
“But how did they know? How did you coordinate?”
“We talked. We knew each other’s situations. When someone needed help, it was known. And people responded.”
Leila found this unsatisfying. The mechanism seemed insufficient to explain the scale of the ERR response. Tens of thousands of volunteers, operating across dozens of cities, delivering aid to millions of displaced people—surely this required more than informal knowledge and cultural tradition?
But the more she studied the data, the more she recognized that her mother was right. The scale emerged from the aggregation of local relationships. Each volunteer knew their neighborhood. Each neighborhood connected to adjacent neighborhoods. The network grew not through central planning but through the repeated invocation of the nafeer primitive.
In El Geneina, as the massacre began, nafeer collided with genocide.
The RSF entered the city from multiple directions on June 9, 2023. The conventional military forces—the SAF garrison, the local police—were overwhelmed within hours. What followed was not combat but hunting.
But even in the hunting, nafeer persisted.
A woman named Halima Musa was hiding with her children in a storage shed behind her house when she heard knocking. She froze—the RSF had been going door to door, and knocking was the prelude to death. But the knocking was followed by a whispered voice she recognized: her neighbor, Yusuf.
“There’s a way out,” he said through the door. “A group is gathering. We can reach the wadi.”
Halima made a choice in that moment—the choice that nafeer demanded. She trusted her neighbor. She gathered her children. They joined a group of perhaps thirty people, moving through alleys and backyards, guided by residents who knew the terrain better than the RSF did.
They reached the wadi—a seasonal riverbed, dry this time of year—and followed it west, toward Chad. The journey took three days. Along the way, others joined them. Villagers who saw them passing offered water, food, directions to safer routes. The nafeer network extended into the countryside, a web of mutual aid that the genocide machine had not yet severed.
Halima reached Adré, across the border, on June 12. She had lost her house, her possessions, her husband (who had stayed behind to cover their escape and did not follow). But she had kept her children alive, and she had done so through the primitive that her culture had prepared for exactly this purpose.
The formal NGO system operated on different principles.
International humanitarian organizations had budgets, mandates, protocols. They conducted needs assessments, developed response plans, secured funding commitments. They coordinated through the UN cluster system, with designated lead agencies for each sector: food, health, shelter, protection.
This system was designed for scale. It could mobilize billions of dollars and deploy resources across continents. It had saved millions of lives in previous crises.
But the Sudan war broke the system.
The scale of displacement overwhelmed assessment capacity. By the time a needs assessment was completed, the population had moved again. The security environment prevented access. Aid convoys were attacked; warehouses were looted; staff were killed or evacuated. The bureaucratic timelines—designed for deliberation and accountability—were too slow for a crisis that killed thousands per week.
The ERRs operated where the formal system could not. They were already inside the crisis, already connected to the affected populations, already trusted. They did not need assessment protocols because they knew their neighbors. They did not need coordination mechanisms because they already coordinated through nafeer.
“We’ve become the UN,” Aisha observed in August 2023, during a rare moment of reflection. “Except we actually get things done.”
The observation was uncharitable to the UN, which faced constraints the ERRs did not. But it captured something real: the informal system was outperforming the formal system in the specific context of the Sudan war.
Nafeer had limits.
The primitive worked best in stable communities where people knew each other, where trust had been established through repeated interaction over years. It worked less well among strangers, among displaced populations thrown together by circumstance, among people who had no prior relationship to draw upon.
As the war progressed and displacement increased, these limits became apparent. The ERR networks were strongest in their original neighborhoods, weakest in the refugee camps that sprang up to absorb the displaced. The nafeer primitive assumed community; displacement destroyed community.
Samira confronted this in September 2023, when she was coordinating aid distribution in a camp outside Port Sudan. The camp held forty thousand people from dozens of different communities. They did not know each other. They had no shared history of mutual aid. They were strangers united only by shared catastrophe.
“How do we do nafeer here?” a younger volunteer asked her.
Samira did not have an answer. The primitive required adaptation. The cultural substrate that had sustained her people for centuries needed to be extended to contexts it had never encountered.
She tried different approaches. She organized by original community, grouping people from the same town or region. She established systems for need identification that did not depend on prior knowledge. She trained volunteers in the methods she had developed in Bahri, transmitting the nafeer primitive to new carriers.
Some of it worked. The camp developed its own informal structures, its own patterns of mutual aid. But the efficiency was lower, the trust thinner, the response slower than in established communities.
The nafeer primitive was powerful but not universal. Its power derived from context, and context could be destroyed.
The parallel with the gold primitive was instructive.
Gold was extracted from specific locations—Jebel Amer, other mining sites. Control of those locations determined access to the resource. The RSF’s power depended on maintaining that control.
Nafeer was extracted from specific communities—neighborhoods, villages, extended family networks. The strength of those communities determined the intensity of the resource. Displacement and genocide were, in this sense, attacks on the nafeer infrastructure. They destroyed the communities that sustained the primitive.
This was not accidental. The genocide was designed, among other things, to break the capacity for collective resistance. Kill the community leaders. Scatter the populations. Sever the relationships that enabled coordination. The gold primitive was self-sustaining through positive feedback; the nafeer primitive was vulnerable to targeted destruction.
“They’re not just killing people,” Leila observed, her analytical frame finally grasping something her mother had known intuitively. “They’re killing the possibility of people helping each other.”
The genocide was not merely murder. It was an attack on the social infrastructure that made survival possible. The displacement was not incidental to the violence; it was an objective. Refugees in camps, strangers to each other, could not mobilize nafeer as effectively as established communities.
But the nafeer primitive had a property that gold lacked: it could be reconstructed.
Gold, once extracted, was gone. The deposits depleted over time. The Jebel Amer mines would eventually run out, and no amount of effort could replace what had been removed.
Nafeer, destroyed in one place, could be rebuilt in another. The primitive was not material; it was cultural. It existed in people, in relationships, in practices. People could be displaced, but they carried the primitive with them. Relationships could be severed, but new relationships could form. Practices could be interrupted, but they could also be transmitted.
In the camps, over months and then years, the nafeer networks reconstituted themselves. Strangers became acquaintances became neighbors became community. The shared experience of survival created new bonds. The ERR volunteers, trained in one context, adapted their methods to new contexts.
By late 2024, the camps had their own ERR structures—different from the neighborhood networks in Khartoum or the village networks in Darfur, but recognizably drawing on the same primitive. The culture had survived the destruction of its original substrate.
“We are still here,” Samira said, in a speech to ERR coordinators in October 2024. “They broke our homes. They scattered us across three countries. They killed our friends and our families. But they did not kill nafeer. They cannot kill nafeer. Because nafeer is not a thing you can burn. It is a thing we carry in ourselves.”
This is the fundamental asymmetry between the two primitive resources:
Gold is finite. Nafeer is renewable.
Gold requires specific geography. Nafeer requires only people willing to help each other.
Gold funds destruction. Nafeer funds reconstruction.
The war in Sudan was, among other things, a competition between these two primitives. The RSF had gold and the capacity for violence that gold enabled. The ERRs had nafeer and the capacity for collective response that nafeer enabled.
In purely material terms, gold was more powerful. It purchased weapons that killed. Nafeer could not stop bullets.
But in terms of persistence, nafeer had advantages. The gold would eventually run out or become inaccessible. Nafeer would persist as long as Sudanese people remembered how to call for help and how to answer that call.
The automaton processed both primitives. The gold funded the genocide machine. The nafeer funded the resistance network. Neither could eliminate the other. Both would continue operating as long as the war continued.
Part I was ending. The primitives were established.
Now they would begin to compose.
Definition: In cellular automata, the adjacency rule determines which cells influence which others. A cell’s next state depends not only on its current state but on the states of its neighbors. Adjacency defines vulnerability: what is near the fire burns.
Maps lie by implication.
A map shows El Geneina as a point, or perhaps a small circle, in the western corner of Sudan. The scale suggests isolation—a city far from centers of power, separated from the capital by a thousand kilometers of desert and scrub. The map implies that what happens in El Geneina is contained, local, unconnected to the larger system.
But the map does not show adjacency.
El Geneina was adjacent to RSF strongholds across Darfur. It was adjacent to Chad, where refugees could flee. It was adjacent to ethnic communities—Masalit neighborhoods beside Arab settlements—whose proximity transformed demographic distribution into targeting data. It was adjacent, through kinship and communication, to diaspora networks in Cairo and London and Minneapolis.
Adjacency is not geography. Adjacency is connection—the channels through which influence flows.
Colonel Yusuf understood this as he fortified El Fasher in June 2023. The city’s physical location was militarily defensible: high ground, limited approach routes, substantial garrison. But the city was adjacent to everything the RSF wanted to destroy. It was the last SAF stronghold in Darfur, surrounded by territory the RSF controlled or contested. It was adjacent to populations the RSF intended to eliminate.
“We’re not isolated,” he told Ismail, studying the map in his command post. “We’re connected to every massacre they’re planning. When they finish with El Geneina, they’ll come here.”
The adjacency determined the sequence. The RSF could not leave El Fasher unconquered while it operated elsewhere in Darfur. The garrison was a potential threat to their supply lines, their administrative control, their project of ethnic consolidation. The connection made the city a target.
The cellular automaton model clarifies adjacency dynamics.
Imagine Sudan as a grid of cells. Each cell represents a locality—a village, a neighborhood, a camp. Each cell has a state: peaceful, contested, SAF-controlled, RSF-controlled, destroyed, evacuated.
The transition rule is: a cell’s next state depends on its current state and the states of its adjacent cells.
Consider a peaceful village next to an RSF-controlled zone. The adjacency creates pressure. RSF forces move through the area. They demand support: supplies, recruits, information. Resistance is punished. Compliance is rewarded, at least temporarily.
The village’s options narrow. It can comply, becoming effectively RSF-controlled without formal conquest. It can resist, risking destruction. It can evacuate, becoming depopulated. Each option changes the cell’s state.
The adjacency also propagates. A village that becomes RSF-controlled creates new adjacency for its neighbors. The pressure extends outward, cell by cell, like infection spreading through tissue.
This is how territory changes hands in Sudan’s war—not through grand military campaigns but through the accumulation of local transitions, each cell responding to the states of its neighbors.
Ibrahim experienced adjacency on June 12, 2023.
He and Khadija had been traveling for three days, moving west toward Chad. The route took them through villages they had never visited, communities whose adjacency to El Geneina made them part of the same crisis.
In the first village, they found help. The residents had not yet been attacked, but they knew what was coming. They shared food with the refugees, offered directions, warned of roads where RSF checkpoints had been established.
“Go through the wadi,” an old man advised. “It’s dry, but there’s cover. The technicals can’t follow you there.”
This village was adjacent to the massacre but not yet consumed by it. Its state was contested—the outcome uncertain, the residents calculating their options.
In the second village, they found evacuation. The houses were empty, the livestock gone, the wells abandoned. Whoever had lived here had made their calculation and departed. The adjacency to violence had produced preemptive flight.
Ibrahim and Khadija rested in an empty house, eating from their dwindling supplies. The silence was absolute. Even the birds seemed to have fled.
In the third village, they found destruction.
The houses were burned. Bodies lay in the streets—men, mostly, killed by gunfire. The smell was what Ibrahim remembered afterward: the specific combination of smoke and decay that marked mass killing. He had never smelled it before. He would never be able to unsmell it.
They did not stop. They passed through quickly, not looking too closely at the bodies, not wanting to recognize faces. The village’s state had transitioned to destroyed. The adjacency to RSF-controlled territory had produced the predictable outcome.
Chad was still thirty kilometers away.
The adjacency rule had a temporal dimension as well.
What happened yesterday influenced what happened today. The state of a cell depended not only on the current states of its neighbors but on the history of those states—the trajectory of change, the pattern of escalation.
Maryam observed this in Khartoum. The neighborhoods that fell first were those adjacent to RSF positions at the start of the war. The neighborhoods that held longest were those with the greatest distance from initial RSF control.
But adjacency shifted as the front lines moved. A neighborhood that was safe in April became contested in May, RSF-controlled in June. The trajectory was predictable if you understood the adjacency dynamics.
By September 2023, she could predict which areas would fall next by examining the map. The RSF advanced along adjacency paths, consuming territory in sequences determined by connection. The areas adjacent to already-controlled zones were always next.
“It’s like a spreading stain,” she told another nurse during a brief conversation between surgeries. “It moves outward from the center. Everything it touches changes color.”
The metaphor was imperfect—the spread was not uniform, not organic in the simple sense. The RSF made decisions about where to attack and when. But the decisions followed a logic that adjacency made comprehensible. They attacked what was nearest to what they already held. They consolidated before expanding. They let adjacency determine sequence.
The counter-system also propagated through adjacency.
The ERR networks spread from neighborhood to neighborhood, each new node connecting to existing nodes. A volunteer in one area would contact a friend in an adjacent area. The friend would begin organizing. The network grew along paths of social connection.
Aisha mapped this growth in Khartoum. In April, ERR nodes existed in perhaps thirty neighborhoods. By June, the number had doubled. By September, ERR presence was nearly universal—every neighborhood had some form of organized response.
The spread followed adjacency patterns. Neighborhoods with strong prior organization (Resistance Committees, professional associations, religious congregations) developed ERR nodes faster. These nodes then catalyzed organization in adjacent neighborhoods through direct contact and example.
“It’s the same dynamic,” Leila observed, comparing her data on ERR spread to data on RSF territorial control. “Both systems propagate through adjacency. The difference is what they propagate.”
The RSF propagated violence. The ERRs propagated capacity. The adjacency rules were mathematically similar even though the content was morally opposite.
Some adjacencies were deliberately severed.
The RSF’s tactic of establishing checkpoints was, among other things, an attempt to control adjacency. A checkpoint interrupted the connection between one area and another. It filtered who could pass and who could not. It made some adjacencies lethal.
The checkpoints around El Fasher served this purpose. The city was surrounded—not by a continuous line of troops, which would be impossible to maintain, but by control of the roads. The adjacency to the outside world was choked.
Yusuf analyzed the checkpoint positions with military precision. The RSF controlled the main roads to Nyala, to Khartoum, to Chad. Secondary routes remained open, but secondary routes were slower, more dangerous, less capable of sustaining supply. The siege was implemented through adjacency control.
“They’re not trying to storm us,” he told Ismail. “They’re trying to starve us. The checkpoint strategy is an attrition strategy.”
The analysis was correct. The RSF understood that a direct assault on El Fasher would be costly. The garrison was well-defended; the population was hostile; the urban terrain favored defense. But if the adjacencies were controlled, the city would weaken over time. Supplies would dwindle. Population would flee. The siege would succeed without a decisive battle.
This is why siege warfare works: it converts adjacency from connection to constraint.
The diaspora represented a different kind of adjacency—adjacency across distance.
Leila, in London, was not physically adjacent to Sudan. But she was socially adjacent through family, through WhatsApp groups, through the coordination platform she had built. The social adjacency enabled flows of information, money, and influence that the geographic distance would otherwise prevent.
The diaspora’s adjacency was selective. Leila was connected to ERR coordinators, to displaced relatives, to activists and journalists. She was not connected to RSF fighters or to the gold traders in Dubai. The social network had structure; it connected some nodes and not others.
This selective adjacency created asymmetric information flows. Leila knew what was happening in ERR-served communities because she was adjacent to them. She did not know what was happening in RSF-controlled areas because she was not adjacent to anyone there.
“I can see half the country,” she complained to a fellow coordinator. “The half we’re helping. The other half is dark.”
The darkness was a function of adjacency. The flows of information followed the same paths as the flows of solidarity. Where the ERR networks reached, data reached. Where they did not reach, silence.
In El Geneina, as June progressed, the adjacency became fatal.
The RSF’s operation systematically converted the city from contested to destroyed, neighborhood by neighborhood. The sequence followed adjacency logic: they began at the edges, where their control was strongest, and worked inward.
The Masalit neighborhoods were targeted first—they were adjacent to the RSF positions outside the city and composed of the population the RSF intended to eliminate. The Arab neighborhoods were mostly spared, or at least not systematically destroyed. The ethnicity of residents determined the transition rule.
Ibrahim’s neighborhood had fallen early. He and Khadija had escaped before the door-to-door operations began. Others had not been as quick, or as lucky, or as prepared by prior trauma.
By June 15, the peak of the massacre, the city’s state was effectively destroyed. Not every building was burned—the physical infrastructure remained partly intact—but the Masalit population was killed, fled, or hiding. The human system that had made El Geneina a living city was broken.
The adjacency now worked in reverse. The destruction in El Geneina influenced the states of adjacent cells. Villages that had been peaceful became contested as refugees arrived and RSF forces pursued. The horror propagated outward along the adjacency paths.
Khadija crossed into Chad on June 14.
The border was marked by a dry riverbed and a Chadian military post. The post was overwhelmed—thousands of people arriving daily, far more than the infrastructure could process. The Chadians were trying to register arrivals, to establish some order, to prevent the chaos from spreading into their territory.
Ibrahim and Khadija waited in a line that barely moved. The sun was brutal. The water they carried was almost gone. Around them, other survivors shared stories, searching for news of family members, assembling fragments of information about what had happened and to whom.
The adjacency to Sudan remained, even after crossing the border. The news flowed with the refugees. The grief and fear traveled with them. The massacre was not contained by the line on the map; it propagated through the people who escaped it.
“We’re not safe yet,” Khadija said, watching the border with wary eyes. “They could follow.”
She was right to worry. RSF forces had pursued refugees into Chad in the past, violating sovereignty with impunity. The international border was supposed to sever adjacency—to mark the edge of Sudan’s war and the beginning of neutral territory. But borders were social constructs, and social constructs depended on enforcement.
The Chadian military held the border, for now. But the adjacency persisted in every other sense. The refugees were still connected to what they had fled. The trauma was still propagating. The genocide was not contained by geography.
The adjacency rule is deterministic in principle but stochastic in practice.
In principle, the state of each cell is determined by the states of its neighbors according to fixed rules. The automaton evolves mechanically, one generation to the next, the future implied by the present.
In practice, uncertainty pervades. The rules are not perfectly known. The states of neighbors are not perfectly observed. Random events—a decision by a single commander, a road washed out by rain, a checkpoint guard who looks the other way—alter trajectories in ways that cannot be predicted.
This was Ibrahim’s experience throughout June 2023. Each day brought new information, new contingencies, new calculations. The transition rules operated, but their operation was mediated by endless local variation. He survived not because he computed the optimal path but because he made reasonable guesses under conditions of profound uncertainty.
The automaton model captures essential dynamics—the propagation through adjacency, the state transitions, the patterns that emerge—but it cannot capture the human experience of living inside the system. The cells do not know they are cells. They experience only their local situation, their immediate neighbors, their desperate choices.
Ibrahim had reached Chad. He was alive. Khadija was alive. But the adjacency would continue to shape their future. The war was not finished with them merely because they had crossed a line.
Part II had begun. The transition rules would now be examined in detail.
Definition: The ethnic rule is a transition function that selects targets based on identity. IF(ethnicity ∈ target_set) THEN(apply_violence). The rule transforms demographic data into killing lists. It is not the only rule in the genocide system, but it is the defining rule—the one that distinguishes genocide from ordinary war.
Bashir learned the ethnic rule without anyone explaining it.
In 2003, during his first operations with the Janjaweed, the targets were clear: Fur villages, Masalit villages, Zaghawa villages. The rule was not stated as an abstraction. It was embodied in the operations themselves. You went to certain places and did certain things to certain people. The pattern was obvious after two or three operations.
The commanders never said “we are targeting people based on ethnicity.” They said “these villages support the rebels” or “these areas harbor terrorists” or “the enemy is here.” But the villages were always Fur, always Masalit, always Zaghawa. The ethnic correlation was perfect.
Bashir was not stupid. He understood what the correlation meant. But understanding did not change his behavior. The orders came; he followed them. The ethnic rule operated through him without requiring his conscious endorsement.
By 2023, when the war against the SAF began, the ethnic rule remained active but the context had shifted. In Khartoum, the fighting was initially military: RSF versus SAF, armed forces against armed forces. The ethnicity of the combatants was mixed on both sides.
But in Darfur, the old rule reasserted itself. The RSF’s operations targeted the same populations they had targeted twenty years before. The same villages burned. The same ethnic groups fled. The pattern replicated with such fidelity that survivors of 2003 recognized it immediately.
Khadija had said it on April 15, the first day of the war: “They’re not here for the rebels. They’re here for us.”
The ethnic rule had never stopped operating. It had only been in abeyance, waiting for conditions that would activate it again.
The rule’s logic was not complex.
In computational terms: FOR each individual, EVALUATE ethnic identity. IF identity matches target category, THEN apply procedure (separation, execution, etc.). ELSE vary treatment based on secondary criteria.
The evaluation step required identification mechanisms. How did you determine someone’s ethnicity in a country where ethnic groups had intermarried for centuries, where appearance was not always diagnostic, where people might claim different identities in different contexts?
The RSF used multiple methods:
Names. Certain names correlated with certain ethnicities. A man named Abdallah might be Arab or non-Arab, but a man named Khamis or a woman named Atiya was almost certainly Masalit. Names were the quickest screen.
Appearance. Skin tone, facial features, body type varied somewhat between groups. The differences were statistical, not absolute—many individuals could not be classified by appearance alone. But combined with other cues, appearance contributed to evaluation.
Language. Dialect, accent, vocabulary revealed background. Even Arabic speakers from different regions had distinctive patterns. The Masalit language, the Fur language, the Zaghawa language were immediate identifiers when spoken.
Location. Where you lived, where you came from, where your family was from. Geography correlated with ethnicity because settlement patterns were historically clustered. Being from a Masalit village was strong evidence of Masalit identity.
Social network. Who you knew, who your relatives were, who vouched for you. The RSF sometimes asked prisoners to identify others. Being named by a person under interrogation could mark you for execution.
The methods were imperfect. Misidentification occurred. Arab individuals were sometimes killed; non-Arab individuals sometimes survived through misclassification. But the system did not require perfection. It required only sufficient accuracy to achieve the operational goal: the elimination of target populations as a coherent presence.
Ibrahim was Masalit.
His identity was not hidden or ambiguous. His name was Masalit. His appearance was recognizable to anyone from Darfur. His language—when he spoke the mother tongue he had learned from his grandparents—was unmistakably Masalit. His village of origin, his extended family, his place in the social network of El Geneina—all of it pointed to the same classification.
The ethnic rule made him a target by definition. Not for anything he had done. Not for any allegiance to rebels or sympathy for the SAF. Simply for being what he was: a Masalit man of military age living in West Darfur when the RSF came.
He understood this with perfect clarity as he fled. The RSF did not care about his politics. They did not care about his profession as a teacher. They did not care about the testimonies he had collected, documenting their crimes. They cared only about his ethnicity.
“They would kill me for my name,” he told Khadija during their flight. “For my face. For the language my grandmother spoke.”
“Yes,” she said. There was nothing more to say.
The ethnic rule reduced him to a category. All his particularity—his education, his beliefs, his decades of teaching children to love Arabic literature—was irrelevant. The rule saw only the classification.
Maryam occupied a different position in the ethnic taxonomy.
She was from the Nile Valley, from the ethnic groups that dominated Sudanese politics and economy. In the RSF’s classificatory schema, she was not a target. She was potentially an ally, or at least a neutral—someone whose ethnicity did not automatically mark her for death.
This relative safety had enabled her to continue working in Khartoum through months when others fled. The hospital served everyone—SAF soldiers, RSF fighters, civilians of all backgrounds. The medical primitive operated without ethnic selection.
But she was not blind to what the ethnic rule meant for others.
In June 2023, during a brief period when the hospital received refugees from Darfur, she treated patients whose injuries told the story. Gunshot wounds from executions. Machete cuts from separations. Burns from villages set alight. The pattern was consistent: the patients were Masalit, Fur, Zaghawa. The perpetrators were identifiable by what they had done to whom.
“Why?” she asked a colleague, an older doctor who had worked through previous crises. “What is the logic?”
The doctor looked at her with tired eyes. “You’re from Khartoum. You don’t know the history.”
“I know there was a genocide in Darfur. I know the Janjaweed—”
“The RSF is the Janjaweed. Different name, same people, same ideology. They believe they are Arabs, that Darfur belongs to Arabs, that the Fur and Masalit and Zaghawa are obstacles to be removed.”
“But that’s insane. We’re all Sudanese.”
“Yes. And yet.”
The ethnic rule did not care about the rationality of its premises. It cared only about its execution. The belief system that motivated it might be historically contingent, ideologically constructed, empirically false. But the rule operated regardless.
The ethnic rule interacted with other rules in the genocide system.
The adjacency rule determined where violence occurred. The ethnic rule determined who was targeted within those locations. The two rules composed: violence propagated through adjacency to locations, and within those locations, the ethnic rule selected victims.
The separation rule depended on the ethnic rule for its targeting. Men were separated from women, but not all men—Arab men were usually exempt. The ethnic identification had to occur before the separation could be applied.
The execution rule likewise depended on prior ethnic classification. Those marked as targets by the ethnic rule were routed to execution. Those not marked might be looted, might be displaced, might be threatened—but the systematic killing was reserved for those the ethnic rule identified.
If non-target ethnicity: variable treatment—coercion, displacement, but not systematic killing.
The ethnic rule was the branching point, the conditional that determined which subsequent procedures applied.
Some people fell between classifications.
Children of mixed marriages. People from ethnic boundaries where categories blurred. Individuals who had lived for decades in communities other than their birth community, who spoke languages not native to their parents, who had married across ethnic lines.
The RSF’s ethnic rule handled ambiguity through multiple mechanisms:
Default to death. When uncertain, classify as target. The error of killing a non-target was less costly to the RSF than the error of sparing a target.
Social identification. If physical and nominal identification was inconclusive, ask others. The community would know who belonged and who didn’t. Under threat, people sometimes informed on their neighbors.
Self-identification. Force the individual to declare. Those who claimed target ethnicity were treated accordingly. Those who claimed non-target ethnicity were sometimes believed, sometimes not.
The ambiguity created moral horror beyond the straightforward horror of ethnic killing. People were forced to choose: claim a false identity and perhaps survive, or claim the true identity and die. Neighbors were forced to inform or risk being classified as targets themselves.
The ethnic rule was not only a mechanism for selection. It was a mechanism for moral destruction—for making people complicit in their own targeting or the targeting of others.
In Adré, across the border in Chad, Ibrahim met a man named Salih.
Salih was Masalit but had lived most of his adult life in Khartoum. He had married a woman from the Nile Valley. His children were mixed. His accent, after thirty years in the capital, was more Khartoumi than Darfuri.
When the war began, he had tried to stay neutral. The RSF controlled his neighborhood; they knew his wife’s family; they had not immediately targeted him. He thought he could wait out the conflict.
Then the purges began.
“They came to the neighborhood and asked who was from Darfur,” Salih told Ibrahim. “Someone gave them names. My name was on the list.”
“Your wife’s family couldn’t protect you?”
“They tried. They said I was family now, that my children were Arabs. The RSF didn’t care. They said Darfuri blood is Darfuri blood. It doesn’t matter who you marry.”
Salih had escaped with his wife and children. They had made it to Chad through a route he would not describe—it involved bribes, hiding, and luck. His wife’s family had disowned her for staying with him. She sat nearby as he talked, holding their youngest child, her face expressing nothing.
The ethnic rule had overridden every other connection. Family ties, community integration, linguistic assimilation—none of it mattered. The rule saw only the fundamental classification, and the classification determined fate.
The counter-system operated without an ethnic rule.
The ERR networks did not select beneficiaries by ethnicity. The nafeer primitive that animated them was inherently universalist: you helped your neighbor because they were your neighbor, not because they belonged to a particular group.
In practice, this meant that ERR volunteers often helped people their community might have considered enemies. Arab ERR workers delivered aid to Masalit refugees. Masalit ERR workers helped displaced families from Arab tribes that had supported the RSF.
This was not always easy. The hatred was real; the crimes were documented; the temptation to retaliate through selective assistance was present. But the ERR ethos explicitly rejected ethnic selection.
“We are not them,” Samira told her volunteers in Khartoum. “If we help only our own, we become them. The whole point is that we help everyone.”
The statement was ideological, but it was also strategic. The strength of the ERR networks came from their legitimacy—from the trust that communities placed in their impartiality. If the ERRs were seen as ethnic organizations, their reach would shrink. If they maintained universalism, they could operate anywhere.
The ethnic rule was a transition function that narrowed. The nafeer primitive was a transition function that expanded. The two systems competed not only in material terms but in the logic of their operations.
Governor Khamis Abakar of West Darfur tried to articulate this on June 15, 2023.
He was Masalit—a political leader in a region where ethnic identity was politicized. But he had built coalitions across ethnic lines. His government included Arabs. His appeals for peace were directed at all communities.
That morning, as El Geneina burned around him, he spoke to Al Jazeera. The video would become one of the most viewed documents of the genocide.
“What is happening is not war,” he said. “It is ethnic cleansing. They are killing people based on their ethnicity. This is genocide.”
He named what he saw. The ethnic rule, operating in plain sight, documented by its own victims.
Hours later, the RSF captured him. The details of his death were disputed—some accounts said he was executed immediately, others that he was tortured first. But the outcome was not disputed. The man who had named the ethnic rule was eliminated by the ethnic rule.
His death was itself a message, a transition signal to everyone watching: even those with political protection, with international visibility, with coalitions across ethnic lines—even they were not safe from the ethnic rule.
The ethnic rule was learned behavior.
Children were not born knowing which categories were targets. They absorbed the classifications through exposure, through language, through the accumulated prejudices of their communities. The rule was transmitted across generations, refined through practice, institutionalized through repetition.
Bashir had learned it as a young man. He had taught it, implicitly, to the soldiers he trained. The rule propagated through human substrate, replicating with high fidelity, occasionally mutating but maintaining its essential structure.
The counter-rule—the universalist principle of the ERRs—was also learned behavior. Samira had learned it from her grandmother, refined it through the 2019 revolution, transmitted it to the volunteers she organized. It propagated through different substrate, through communities that had cultivated different values.
The war in Sudan was, at one level, a competition between these two learned rules. The ethnic rule demanded division, demanded classification, demanded death for those classified wrong. The universalist rule demanded solidarity, demanded assistance, demanded help for all regardless of category.
Neither rule was natural in the sense of being biologically inevitable. Both were cultural constructions, products of history and choice. But once learned, both operated with the automaticity of deeply ingrained patterns.
The ethnic rule was winning on the ground. It had weapons, funding, organizational infrastructure. It could kill faster than the universalist rule could save.
But the universalist rule had survival capacity. It could persist through the killing, could reconstitute itself in the camps, could transmit itself to new carriers. The ethnic rule required continuous application of force; the universalist rule required only continuous willingness to help.
The automaton continued processing. The rules continued operating. The outcome remained undetermined.
Definition: The siege rule is a temporal attrition function. WHILE(garrison holds AND supplies > critical threshold), increment time, decrement supplies, apply pressure. The siege transforms time into a weapon. It is patient violence, mechanized waiting, the mathematics of exhaustion.
Colonel Yusuf arrived in El Fasher on June 20, 2023.
The city was not yet besieged—not formally, not completely. The roads to Nyala and Khartoum were still contested. Supplies still reached the garrison through convoys that ran a gauntlet of ambushes. The situation was dire but not yet terminal.
He assessed the position with the analytical detachment that three decades of military service had taught him. The garrison: 4,000 soldiers, well-armed by Sudanese standards, motivated by the knowledge that surrender was not an option for them or their families. The supplies: ammunition for perhaps six months of sustained combat, fuel for three months, food for two months if rationed carefully. The population: approximately 1.5 million, of whom perhaps 800,000 were displaced persons from elsewhere in Darfur, crowded into camps around the city.
“We can hold,” he told his staff at the first briefing. “The question is for how long, and at what cost.”
The question haunted him over the subsequent months. The siege tightened incrementally—a checkpoint established here, a supply route cut there. Each day brought small deteriorations that accumulated into catastrophe.
By December 2023, the last overland route was closed. The airport remained operational—barely—allowing occasional flights with critical supplies. But air supply could not sustain 1.5 million people. The mathematics was inexorable.
The siege rule has ancient roots.
Starvation as a weapon predates gunpowder, predates writing, predates settled civilization. The logic is simple: a population requires inputs to survive. Control the inputs, and you control the population. The besieger need not storm the walls; the besieger need only wait.
The RSF understood this logic intuitively. They had neither the capability nor the inclination for direct assault on El Fasher. The garrison was strong; the terrain favored defense; casualties would be high. Better to encircle and wait.
Their strategy was sophisticated in execution if not in conception. They established control of the major roads, not through continuous positions—which would stretch their forces too thin—but through mobile checkpoints that could intercept convoys. They attacked the airport periodically, not to capture it but to make flights dangerous. They cut the water and electricity infrastructure when they could reach it.
The pressure was calibrated. Too much, and the garrison might attempt a breakout in desperation. Too little, and the siege would not progress. The RSF sought the optimal level—enough pressure to erode capacity, not enough to provoke a decisive response.
This is the mathematics of siege warfare: finding the rate of attrition that maximizes long-term damage while minimizing short-term risk.
Maryam transferred to El Fasher in October 2024.
The decision was not hers. The Ministry of Health—what remained of it, operating from Port Sudan—was attempting to redistribute medical personnel to where they were most needed. El Fasher had desperate need and insufficient staff. Maryam had experience and expertise.
She flew in on one of the last regular flights, a small cargo aircraft that dodged RSF anti-aircraft fire on approach. The landing strip was cratered; the pilot navigated between the holes with practiced skill.
What she found was a hospital operating on the edge of collapse.
The South Hospital had been the main medical facility in El Fasher before the war—100 beds, reasonable equipment, adequate staff. Now it served as the primary healthcare provider for over a million people. The beds were full three patients to each. The equipment was failing, and replacement parts were unavailable. The staff worked twenty-hour shifts until they collapsed.
“Welcome to the siege,” Dr. Fatima, the hospital director, told her on the first day. “Everything you know about medicine, forget. Here we practice triage and prayer.”
Maryam adapted. The principles she had learned in Khartoum’s chaos translated to El Fasher’s systematic disaster. Count what you have. Prioritize ruthlessly. Document everything.
The documentation acquired new importance as the siege continued. The hospital became a record of attrition—the injuries, the diseases, the deaths that accumulated as supplies dwindled.
The siege rule operated through multiple mechanisms simultaneously.
Food scarcity. The civilian population of El Fasher depended on food brought from elsewhere—the city could not feed itself. As convoys stopped, reserves depleted. Prices rose exponentially. By mid-2024, a sack of grain cost ten times its prewar price. By early 2025, it cost fifty times.
Fuel scarcity. The generators that powered the hospitals, the water pumps, the communications equipment—all ran on diesel. Diesel came by truck, and trucks came through roads the RSF controlled. The garrison prioritized fuel for military operations. Civilian facilities went dark.
Medical supply scarcity. Drugs, bandages, surgical equipment, everything consumable in medicine. The supply lines were cut; the supplies depleted; the patients suffered and died from conditions that would have been treatable in normal circumstances.
Water scarcity. The water system required electricity to pump. As electricity became intermittent, water became intermittent. People waited in lines for hours at the few functioning wells. The water they obtained was often contaminated.
Each scarcity amplified the others. Malnutrition weakened immune systems, increasing disease. Disease increased demand for medical care. Lack of medical care increased death rates. Death rates increased desperation, which increased conflict over the remaining resources.
The siege rule was a system, not a single mechanism. Its components interacted to produce effects greater than any component alone.
Yusuf calculated constantly.
He maintained a ledger of supplies—ammunition, fuel, food, medicine—updated daily with consumption rates and remaining stocks. The numbers defined the horizon of possibility.
At current consumption, the garrison could sustain combat operations for another eight months. But “current consumption” assumed a static situation. Increased RSF pressure would accelerate consumption. Reduced supply flights would accelerate depletion. The horizon was not fixed; it receded or approached depending on conditions.
“We’re not defending a city,” he told Ismail during one of their planning sessions. “We’re managing a controlled collapse. The question is whether we can extend the collapse long enough for something to change.”
“What might change?”
“I don’t know. That’s the problem. We’re hoping for rescue, but we don’t know what form the rescue might take. Maybe the SAF recaptures territory elsewhere and relieves us. Maybe international pressure forces a ceasefire. Maybe the RSF runs out of resources before we do.”
“Or maybe nothing changes, and we run out of everything.”
“Yes. That’s also possible.”
The siege imposed this existential uncertainty. Each day that passed was a day closer to depletion—but also a day closer to potential salvation. The mathematics provided no certainty about which limit would be reached first.
The civilian population experienced the siege differently from the garrison.
Soldiers had rations—reduced, but existent. They had access to the remaining fuel, the remaining medicine, the remaining capacity. The military apparatus, even in deterioration, allocated resources to its own.
Civilians had nothing but what they could obtain through their own efforts or the ERR networks.
The ERR in El Fasher operated under impossible constraints. They ran community kitchens when food was available. They maintained water distribution points when pumps worked. They organized medicine sharing, ensuring that the scarce antibiotics and painkillers reached those who needed them most.
But the constraints were impossible. More need than resources. More hunger than food. More sickness than medicine. The ERR volunteers worked themselves to exhaustion and still could not meet the demand.
Maryam observed this from the hospital. The patients who arrived were increasingly malnourished—children with swollen bellies and stick-thin limbs, adults who had lost half their body weight, elderly people whose bones showed through papery skin.
“Starvation is a medical condition,” Dr. Fatima told her. “We treat it with food. But we don’t have food.”
The siege rule made medical practice absurd. They diagnosed hunger and prescribed hunger relief that did not exist. They documented malnutrition deaths knowing that documentation changed nothing.
The RSF’s siege was punctuated by violence.
They did not simply wait. They attacked periodically—probing assaults on the perimeter, artillery barrages on the city center, drone strikes on infrastructure. Each attack served multiple purposes: testing defenses, consuming ammunition, terrorizing the population.
The drone strikes were particularly effective. The RSF had acquired surveillance and attack drones—from where was unclear, but the UAE was suspected. The drones could loiter for hours, watching, waiting, then striking with precision at a generator, a water tank, a gathering of people.
The hospital was hit in March 2025.
Maryam was in the surgical wing when the explosion occurred. The impact was in the administrative building—the part of the hospital that contained records, administrative staff, the director’s office. The drone had struck during a meeting.
Dr. Fatima was killed instantly. Twelve other staff members died. The records room burned.
Maryam survived because she was elsewhere. The randomness of it—that she had been called to a surgery rather than attending the meeting—saved her life.
“Why the hospital?” she asked a military officer who came to inspect the damage. “We treat everyone. We’re neutral.”
The officer looked at her with an expression that combined pity and contempt. “The hospital supports the garrison’s morale. It treats wounded soldiers. It maintains civilian capacity to endure. From the RSF’s perspective, destroying it advances the siege.”
The logic was cold but accurate. The siege rule did not recognize neutrality. Everything that enabled the besieged to continue resisting was a target.
The 920 days accumulated.
Day 100: still receiving occasional supply flights. Morale uncertain but not collapsed. ERR networks functional.
Day 200: supply flights halted after two aircraft shot down. Rationing intensified. First outbreaks of cholera in the camps.
Day 300: fuel exhausted except for emergency reserves. Hospital operating on generator power, two hours daily. Malnutrition deaths accelerating.
Day 500: population visibly reduced—some fled, some died, some evacuated in brief humanitarian corridors. Those remaining were those who could not leave.
Day 700: artillery barrages daily. Perimeter breached multiple times, retaken at cost. Ammunition running low. Trading space for time.
Day 900: final preparations. Distribution of remaining supplies. Destruction of sensitive documents. Acknowledgment that the end was approaching.
Day 920: October 26, 2025.
The mathematics had reached its conclusion. The siege rule had operated for 920 days, and on day 920, the remaining variables dropped below the critical threshold.
The RSF launched their assault at dawn.
Yusuf had known the end was coming for weeks.
The supply situation was terminal. The remaining ammunition would support perhaps one more day of sustained combat. The fuel was gone—the vehicles immobile, the generators silent. The food would last another week, but a week without ammunition was a week without defense.
He had considered options. A breakout attempt—futile, given the encirclement. A negotiated surrender—the RSF had made clear that surrender would mean death for the garrison commanders, whatever terms might be offered. A final stand—tactically hopeless, but perhaps preferable to the alternatives.
In the end, events overtook planning.
The RSF breached the northern perimeter at 06:00. By 10:00, they controlled a third of the city. By noon, Yusuf’s command post was surrounded.
He burned the last documents. He ensured that Ismail and several other junior officers had escape routes identified—they might make it to the desert, might reach Chad, might survive. He himself would not try to escape. The commander stayed with his command.
The RSF captured him at 11:17. He did not resist. There was nothing left to resist with.
The siege rule produced its output: a captured city, a destroyed garrison, a population delivered to the mercy of those who had besieged them.
But the rule had not operated alone. The siege had been sustained by the gold primitive, which funded the RSF’s patient attrition. It had been resisted by the nafeer primitive, which kept the civilian population alive far longer than pure calculation would have predicted. It had been documented by witnesses like Maryam, who would carry the record even after the hospitals burned.
The siege was over. What followed was massacre.
The rules of genocide—ethnic selection, separation, execution—now applied to a population that had been softened by 920 days of starvation and bombardment. The siege had been the setup; the massacre was the execution.
Maryam was in the hospital when the RSF entered on October 26. She hid for fourteen hours, listening to the sounds above—gunfire, screaming, silence. When she emerged, the hospital was empty except for the dead.
The siege rule had completed its operation. The cellular automaton had processed through 920 generations. The state of El Fasher had transitioned from contested to RSF-controlled to destroyed.
The transition rules continued.
Definition: The separation rule partitions a population into subsets for differential treatment. SORT(population) → {target_subset, non-target_subset, uncertain_subset}. Separation is the prerequisite for selective violence. You cannot kill some while sparing others until you have distinguished some from others.
The separation began with a simple command: “Men on this side. Women and children on that side.”
Bashir had spoken these words hundreds of times. The phrasing varied—sometimes “males over fifteen,” sometimes “anyone who can carry a weapon,” sometimes just a gesture with the rifle barrel. But the function was constant: partition the population by characteristics that correlated with targeting criteria.
The separation served multiple purposes.
First, it identified. Men of fighting age were potential combatants, potential resisters, potential witnesses who might later seek revenge. Separating them made them visible, countable, manageable.
Second, it controlled. A mixed population could coordinate resistance. Separated groups could not easily communicate, could not plan together, could not protect each other. The separation atomized collective capacity.
Third, it terrorized. The moment of separation was the moment when everyone understood what was coming. The women watching their husbands and sons led away knew. The men being led away knew. The knowledge itself was a weapon.
In El Geneina, in June 2023, the separation operated at industrial scale. Neighborhood by neighborhood, the RSF moved through Masalit areas, gathering populations, applying the partition function.
Halima Musa witnessed separation before she escaped.
She was hiding with her children in the storage shed when her husband, Mahmoud, decided to check on the neighbors. He had been gone for ten minutes when she heard the voices—Arabic commands, the sound of people being moved.
Through a crack in the shed wall, she watched.
The RSF had gathered perhaps forty people from the surrounding houses. They were sorting them with quick efficiency. Men pushed to the left, toward a waiting truck. Women and children pushed to the right, held in a loose cluster by two guards.
She saw Mahmoud among the men. He was not resisting—no one was resisting. The rifles made resistance meaningless. He stood with the others, his face showing the blank expression of someone who has understood his situation completely.
The truck filled. More men were brought. When the truck could hold no more, it drove away. She would learn later that trucks like that one went to execution sites outside the city—wadis where the killing could happen away from witnesses, where the bodies could be buried or burned.
The women and children remained. Their fate was different but not better. Some were released after looting. Some were taken to other locations. Some were subjected to violence that Halima could not let herself imagine, even months later in the safety of the refugee camp.
She stayed hidden. Her children stayed silent—even the youngest, only four, seemed to understand that sound meant death. When darkness fell and the RSF moved on, she gathered her children and ran.
She never saw Mahmoud again.
The separation rule had variants.
Gender separation was the most common: men from women. This correlated with the targeting of military-age males as presumed combatants.
Age separation refined the gender split: boys above a certain age were classified with men; elderly men were sometimes spared. The age threshold varied by operation, by commander, by the perceived threat level of the population.
Ethnic separation occurred in mixed areas: Arabs separated from Masalit, Fur, Zaghawa. The ethnic rule operated through separation, identifying targets for the subsequent execution rule.
Occupational separation targeted specific categories: teachers, government workers, community leaders. These were the “articulators”—people capable of organizing resistance or documenting crimes. They were separated for special treatment.
Governor Khamis Abakar was an instance of occupational separation. He was not merely Masalit; he was the chief administrative officer of the region, the person most capable of coordinating response and most likely to be heard by the outside world. His separation and execution was a priority.
Ibrahim, as a teacher who documented testimonies, would have been subject to occupational separation had he been captured. His profession made him doubly dangerous: educated enough to understand, articulate enough to communicate, persistent enough to record.
The counter-system also employed separation, but with opposite logic.
The ERR networks separated populations by need rather than by identity. Who was hungry? Who was sick? Who needed evacuation? The sorting served allocation rather than elimination.
Aisha developed triage protocols in the early months of the war. The protocols were ruthless in their own way—resources were scarce, and not everyone could be helped. But the ruthlessness served survival, not killing.
“We separate because we have to prioritize,” she explained to new volunteers. “A family with young children gets food before a family that can forage. A patient who will die without medicine gets medicine before a patient who will survive without it. We can’t help everyone, so we help those we can save.”
The logic was medical triage applied to humanitarian response. It accepted that choices had to be made, that some would receive less than they needed, that perfect equity was impossible. But the choices served the goal of maximum survival, not selective death.
This was the moral difference between the two separation systems: one sorted for killing, the other sorted for saving. The mechanical operation was similar—partition a population into subsets for differential treatment—but the purpose inverted the meaning.
Tariq, the conscripted engineering student, participated in his first separation in July 2023.
His unit had been assigned to clear a neighborhood in southern Khartoum. The procedure was familiar from training: surround the area, move through systematically, gather the population, separate and process.
He did not pull the trigger on anyone that day. His role was perimeter security—standing at a designated point, rifle ready, ensuring no one escaped the cordon. But he watched the separation happen.
The soldiers doing the sorting were efficient. They had done this before. Their movements were practiced, their commands standardized. “Men here. Women there. Move. Move.”
A family came through his sector—a father, a mother, three children. The father was perhaps forty, dressed in the work clothes of a laborer. The mother carried the youngest child. The older two walked between their parents.
The sorting team reached them. The father was pulled away, pushed toward the cluster of men. The mother cried out, reaching for him. A soldier shoved her back. The children wailed.
Tariq watched. His rifle was in his hands. He could have—what? Intervened? Shot his fellow soldiers? Declared that this was wrong and he would have no part of it?
He did nothing. He watched the father disappear into the cluster of men. He watched the mother collapse, held up by another woman. He watched the children cry.
Later, he would tell himself that there was nothing he could have done. The machinery was too large, too well-organized, too indifferent to individual resistance. One soldier refusing orders would change nothing except his own survival.
But he knew, even as he constructed the justification, that the justification was also a confession. He had participated. He had enabled. He had been a component in the separation system.
The separation rule created information.
Before separation, the population was undifferentiated—a mass of people whose individual characteristics were unknown. After separation, the population was organized—the men here, the women there, the leaders identified, the fighters distinguished from civilians.
This information was operationally valuable. It enabled targeted violence. It enabled selective extraction—taking fighters for execution while using women for other purposes. It enabled documentation of who had been processed, which households had been cleared, what remained to be done.
The RSF maintained records. Not comprehensive records—the chaos of operations prevented that—but sufficient records to track progress. Which neighborhoods were complete. Which had residual population. Which required follow-up.
Leila discovered these records later, when captured RSF documents were smuggled out by survivors. The bureaucratic nature of the genocide was documented in its own filing systems: lists of names, tallies of “processed” households, notes on remaining tasks.
“They’re running it like a project,” she told the diaspora coordination meeting, displaying scanned documents on her screen. “There are checkboxes. There are completion percentages. They’re managing the genocide like it’s a construction project.”
The horror was not only in the killing. It was in the rationalization of killing—the transformation of murder into procedure, atrocity into administration.
Some people escaped separation through disguise.
A man might hide among women, covering himself in a woman’s tobe, hoping to pass initial inspection. A Masalit might claim to be Arab, speaking Arabic without the tell-tale accent, hoping the sorting would misclassify.
These escapes were rare. The separation teams were experienced; they looked for disguises. The physical indicators of gender were usually obvious. The community members forced to identify their neighbors would often reveal the pretenders.
But some succeeded.
A survivor named Ismail—not Yusuf’s aide, a different Ismail—told Ibrahim his story in the Adré camp:
“I was with the men when they sorted us. I knew what was coming. When the guards looked away for a moment, I slipped behind a building and hid in a garbage pile. I stayed there for two days, not moving, barely breathing. The smell kept them from searching too close.”
He had survived by removing himself from the separation system entirely—by becoming invisible to the sorting algorithm, unclassifiable because not observed.
“Why you?” Ibrahim asked. “Why did you think to hide when others didn’t?”
Ismail shook his head. “I don’t know. Maybe because I had seen it before. My brother was killed in 2003, the same way. When I saw them starting to separate us, I remembered what came next.”
Prior exposure to the pattern enabled prediction. Ismail’s brother’s death had been, in a terrible sense, a lesson that allowed Ismail to survive twenty years later.
The separation rule reveals something fundamental about genocide’s logic.
Mass killing is difficult. A population resists. It fights, flees, hides. It is heterogeneous—mixing targets with non-targets, combatants with civilians, those marked for death with those who might be spared.
Separation solves this problem. It transforms a complex, resistant population into organized subsets that can be processed efficiently. It applies the power of categorization—the same cognitive operation that enables all systematic thought—to the project of elimination.
The bureaucrats who designed the Holocaust understood this. The Einsatzgruppen who shot Jews in Eastern Europe learned through practice that separation increased efficiency. The Hutu militias in Rwanda used separation to manage the flow of victims at roadblocks.
The RSF applied the same principle, adapted to Sudanese conditions. The principle was not invented; it was inherited—part of the accumulated knowledge of how to conduct genocide, transmitted through history and example.
Yusuf, in his command post, knew that capture meant separation.
If El Fasher fell—when El Fasher fell—the population would be sorted. The SAF garrison would be separated from civilians. Within the garrison, officers would be separated from enlisted men. Within the officers, Yusuf himself would be separated as the commander.
His category was clear: SAF officer, garrison commander, Zaghawa by ethnicity. Every dimension of his identity pointed to the same classification. The separation would route him to execution with certainty.
He did not fear death, exactly. Thirty-one years of military service had accustomed him to mortality’s proximity. But he feared what his death would mean—the loss of the garrison, the abandonment of the population they had protected, the victory of a force he considered not merely enemy but evil.
“If they take me,” he told Ismail on the night of October 25, “do not try to rescue me. Get out. Get to Chad. Tell the world what happened here.”
“Sir—”
“That’s an order. The important thing is not my survival. The important thing is that someone survives to be a witness. That someone remembers.”
The separation was coming. When it came, those who survived would carry the record. Those who did not survive would become the record—the documentation of what the genocide had accomplished.
On October 26, 2025, at 06:47, the RSF began separation operations in El Fasher.
The pattern was the same as it had been in El Geneina, in Ardamata, in dozens of other locations. The population was gathered, sorted, distributed for differential treatment.
Men to the stadium. Women to the detention centers. Children—the children were sometimes separated from women, sometimes kept with them, depending on the unit and the commander.
Maryam, hiding in the hospital, could hear the operations through the walls. The commands, the screaming, the gunfire that punctuated the sorting. She could not see, but she could reconstruct the sequence from sound.
The separation continued throughout the day. Trucks moved through the streets, collecting the sorted subsets. The stadium filled with men. The detention centers filled with women. The morgues and the mass graves filled with those who resisted or those who were executed immediately.
By evening, the city was sorted. The population of 1.5 million had been partitioned into categories: dead, detained, fled, hiding. The separation was complete.
What remained was the subsequent rules: execution, looting, erasure. The separation had created the conditions for these operations. The algorithm was proceeding.
Definition: The execution rule converts separated targets into eliminated targets. APPLY(lethal_force) TO (target_subset) → dead. The rule is the terminal operation in the genocide sequence. Everything preceding—identification, separation, detention—leads to this. The execution rule is where the automaton’s purpose is finally expressed.
The mathematics of mass killing is simple.
A bullet costs approximately thirty cents. A person can be killed with one bullet if the shot is well-aimed. A fighter with a rifle can kill one person approximately every ten seconds if the targets are lined up, every thirty seconds if they must be chased and cornered.
At this rate, one hundred armed men can kill one thousand people in roughly three hours. Ten thousand people in thirty hours. One hundred thousand people in two weeks, working in shifts.
These calculations are not abstract. They were performed, explicitly or implicitly, by every organization that has conducted genocide. The Nazis calculated the throughput of gas chambers. The Rwandan Interahamwe calculated how many could be killed with machetes before the killing exhausted. The RSF calculated how many fighters were needed, how much ammunition was required, how long the operations would take.
The mathematics does not capture the horror. But it explains the possibility. Genocide is not spontaneous passion; it is organized logistics. The execution rule is implementable because the implementation has been planned.
In El Geneina, June 2023, the execution sites were outside the city.
The wadis—dry riverbeds—provided natural depressions suitable for mass graves. Trucks transported separated men to these locations. The killings occurred away from the main population, reducing witness testimony and enabling volume processing.
Bashir was present at one of these sites on June 17, 2023.
The truck had brought perhaps eighty men—the capacity of the vehicle, packed standing. They were unloaded and arranged in a line along the edge of the wadi. Some were bound; most were not. Binding was unnecessary when armed guards surrounded them and there was nowhere to run.
The commander gave the order. The line of fighters opened fire.
The first volley killed most of them. Some fell immediately, dead before they understood. Others fell wounded, screaming. The fighters walked among the fallen, firing single shots into those who still moved.
The bodies were pushed into the wadi. A front-loader covered them with earth. The truck returned for another load.
Bashir performed his function. He fired when ordered, confirmed kills when required, helped load the next transport. His hands remained steady throughout—the same steadiness he had noted after his first kill at seventeen.
What he thought, what he felt, whether he processed the events as murder or as operation—these questions are unanswerable from outside his consciousness. His behavior was indistinguishable from that of his fellow fighters. The execution rule operated through all of them equally.
The rule had variations.
Mass shooting was the primary method: efficient, requiring only ammunition and coordination. The RSF had sufficient ammunition, supplied through the gold-funded supply chains that ran through Dubai.
Individual execution was used for high-value targets: commanders, officials, witnesses whose deaths served symbolic purpose. Governor Khamis Abakar was killed individually, his death documented and communicated as a message.
Burning was used to eliminate bodies and destroy evidence. Some victims were burned alive; others were burned post-mortem. Fire consumed identifying features, made counting difficult, erased the specificity of the dead.
Starvation was a slower execution method, applied through siege. The death came not from bullets but from the absence of food. The El Fasher siege was, in this sense, a 920-day execution—the gradual application of lethal force through deprivation.
Drowning was used along the rivers—bodies thrown into the water, washing downstream, disappearing into the current. The Nile carried corpses from Khartoum neighborhoods where fighting was heaviest.
Each method had its logic, its efficiency, its signature. The methods could be identified from the bodies, when bodies were found. The forensic evidence told the story of which rule variant had been applied.
The ERR volunteers encountered the output of the execution rule.
In Khartoum, during the periods of shifting control, neighborhoods that changed hands revealed what had happened under RSF occupation. Bodies in houses, in streets, in improvised detention sites. The execution rule left evidence.
Aisha organized the first body recovery operations in August 2023.
“We can’t bury everyone,” she told her team. “But we can document. We can photograph. We can record names when families identify them.”
The work was beyond traumatic. The volunteers moved through scenes of murder, cataloging the dead, noting the locations and visible causes of death. The smell was constant. The flies were everywhere. The psychological cost was incalculable.
But the documentation served a purpose. It created a record. When the RSF later denied what they had done, the photographs and locations and names would contradict the denials.
“Someday there will be trials,” Aisha said, though she did not entirely believe it. “Someday this evidence will matter.”
The execution rule produced bodies. The documentation rule produced records. The two outputs would eventually meet in the forum of justice—or they would not, and the records would simply be the memory of what was done.
Ibrahim, in the Chad camps, collected testimonies of executions.
The survivors described variations of the same pattern. The separation. The transport. The lining up. The shooting. The shallow graves or the burning or the abandonment.
Each testimony added detail. The location of this wadi. The number killed at that site. The commanders who gave orders. The fighters who executed them.
“I want to understand the system,” Ibrahim explained to one survivor who asked why the questions mattered. “Not just what happened, but how it happened. How they organized it. How they decided who to kill and where and when.”
The survivor looked at him with exhausted eyes. “Does it matter? They killed everyone. That’s the system. They killed everyone they could catch.”
“But they didn’t catch everyone. You survived. Others survived. Understanding how the system works tells us where the gaps are. Where people escaped. Where future victims might escape.”
This was the analytical purpose: not merely to record atrocity but to decompose it into components that could be understood and potentially interrupted. The execution rule was terrible, but it was not omnipotent. It had limits, failures, inefficiencies. Understanding the rule meant understanding how to evade it.
The stadium in El Fasher became an execution site on October 26, 2025.
Yusuf was among those brought there, though he did not know the location until he arrived. The transport had been disorienting—a truck with no windows, a journey of perhaps fifteen minutes, arrival at a large enclosed space that he eventually recognized.
The stadium had been built for football matches. It had hosted national team games, school championships, community events. Now it held approximately four thousand men, sitting or standing on the field, surrounded by armed RSF fighters on the bleachers and at the gates.
The executions began in the evening.
They started at one end of the field and moved systematically toward the other. Groups of perhaps fifty men at a time were led to the far goal, where a trench had been dug. The groups were shot into the trench. The next group was led forward.
Yusuf watched from near the center of the field. The mathematics was clear: at the current rate, four to five hours to complete. His position in the sequence gave him approximately two hours.
Two hours to what? To escape? There was no escape. To fight? With what weapons, against how many guards? To pray? He had prayed already, through the 920 days of siege, for outcomes that had not materialized.
He chose to observe. To witness, even knowing that his witness would die with him. To see clearly what was happening, to understand the mechanism that was processing him.
The groups advanced. The shooting continued. The trench filled.
At 21:47, according to his watch—he had kept his watch through the capture, somehow, this small absurdity—the guards came for his section.
He did not die in the stadium.
The execution sequence was interrupted by an explosion—the SAF’s final act, a bombing run that targeted RSF positions throughout the city. One bomb hit near the stadium, close enough to cause chaos among the guards, far enough not to kill the prisoners.
In the confusion, Yusuf ran.
He did not think about running; his body ran. Years of military training asserted itself, overriding paralysis, converting surprise into motion. He was through a gap in the fence before he consciously registered the decision.
Others ran too—perhaps a hundred men from the stadium, scattering in all directions. The RSF fired after them, killing some, missing others. In the darkness and chaos, many escaped.
Yusuf found himself in a drainage ditch, pressing his body into the mud, listening to gunfire and screaming and the sound of vehicles. He stayed there for four hours, until the sounds faded, until the cold became unbearable, until the imperative to move overcame the imperative to hide.
He made his way west. He did not know if Ismail had survived. He did not know who else had escaped the stadium. He knew only that the execution rule had failed to complete its operation on him, that some malfunction in the system had produced a gap through which he had passed.
The automaton was not perfect. Its rules were not inviolable. Even the execution rule could fail.
The failure rate of the execution rule determined the survivor population.
In every genocide, some targets escaped. The mechanisms of escape varied: hiding, disguise, bribery, chaos, luck. The cumulative effect was that genocide never achieved perfect elimination. A remnant survived.
This remnant was dangerous to the perpetrators.
Survivors could testify. They could identify killers. They could contradict the denials that invariably followed genocide. They could seek revenge. They could maintain the memory that the perpetrators wanted erased.
The RSF understood this. Their operations included pursuit of escapees, secondary sweeps to find those who had hidden, execution of witnesses who might later speak. The erasure rule, which followed the execution rule, was partly about eliminating the evidence of killing but also about killing those who had witnessed.
Despite these efforts, survivors persisted. Halima in Chad. Ibrahim in Chad. Yusuf, stumbling west through the desert. Maryam, still hiding in the hospital. Thousands of others who had found gaps in the system, holes in the net, moments of failure in the mechanism.
These survivors would become the carriers of memory. Their continued existence was itself a defeat of the execution rule—a demonstration that the automaton was not omnipotent, that even the most systematic violence could not complete its project.
The execution rule’s operation created a temporal divide.
Before: a population existed. People lived their lives, pursued their projects, imagined their futures. The automaton had not yet processed them.
After: that population was dead, transformed from living subjects into corpses, then into buried remains, then into memory. The automaton had completed its operation.
This divide was absolute in the individual case—a person was alive, then dead, with no intermediate state—but gradual in the collective case. The population was killed over hours, days, weeks. The automaton processed sequentially, not instantaneously.
The gradual processing created the possibility of intervention. Between “before” and “after” was a window in which the execution could be stopped. The siege of El Fasher was, in one sense, a 920-day window—a period in which the international community could have intervened, could have changed the outcome, but did not.
The window closed on October 26, 2025. The execution rule operated. The temporal divide was established.
What remained was the aftermath: the survivors, the evidence, the memory. The execution rule had processed a population, but it had not processed the future. The future remained undetermined, capable of containing justice or impunity, memory or forgetting.
Maryam emerged from hiding on October 27, 2025.
The hospital was silent. The sounds of the previous day—the shooting, the screaming, the commands—had faded. What remained was a silence that was itself a form of sound, an acoustic signature of emptiness.
She found bodies in the corridors. Patients who had been killed in their beds. Staff who had been executed at their stations. The execution rule had operated here too, even in a hospital, even among the wounded and the healers.
She walked through the destruction, counting. Documenting. The pattern-recognition she had developed through months of medical work now applied to a different kind of data.
Thirty-seven dead in the main building. Twelve in the surgical wing. Eight in the morgue, which was darkly appropriate—the dead stored among the dead. Fifty-seven total, all killed within a few hours.
She photographed what she could. She recorded locations, positions, visible injuries. She noted the names of those she recognized—colleagues, patients, people whose lives had intersected with hers during the siege.
The execution rule had produced this output. Her documentation rule would produce its counter-output. The two records would exist in parallel: the killing and the memory of the killing, the operation and the testimony of the operation.
She did not cry. The tears would come later, if they came at all. For now, she had work to do.
The automaton had operated. The survivor had survived. The recording continued.
Definition: The erasure rule eliminates evidence of prior operations. DESTROY(traces) → plausible_deniability. Erasure operates on multiple substrates: physical evidence (bodies, buildings), documentary evidence (records, photographs), testimonial evidence (witnesses). The goal is to make the genocide impossible to prove—or at least difficult enough that proof requires extraordinary effort.
The burning began before the killing ended.
Even as executions continued in El Fasher’s stadium, RSF units were setting fires throughout the city. The targets were systematic: government buildings that might contain records, hospitals that had documented injuries, schools where educated people might have kept notes.
Ibrahim’s school in El Geneina had burned on June 16, 2023. He had not witnessed it—he was already fleeing toward Chad—but other survivors told him. The building where he had taught for twenty years was ash by evening. The records of students, the examination papers, the small library he had accumulated—all erased.
The burning served multiple purposes.
First, it destroyed physical evidence. Bodies could be burned, making identification impossible. Buildings could be burned, eliminating documents that might later be used in prosecution.
Second, it destroyed the possibility of return. A population cannot return to homes that no longer exist. The burning converted displacement into permanence.
Third, it communicated. The smoke rising from a city told neighboring communities what was coming. The erasure was itself a message: this is what happens, and afterward, nothing remains.
The erasure extended to digital space.
Communications blackouts accompanied the major massacres. Cell towers were destroyed or disabled. Internet access was cut. The informational isolation served the same function as physical destruction: preventing documentation, preventing the flow of evidence.
In El Fasher, the blackout began on October 25, 2025—one day before the fall. The last messages from the city were panicked, fragmentary: “They’re coming.” “God help us.” “Tell the world.” Then silence.
Leila, monitoring her coordination platform in London, watched the data streams go dark. One by one, the ERR contacts in El Fasher stopped responding. The map on her screen, which had shown green dots for active nodes, turned gray as the signals ceased.
“We’re blind,” she told the emergency meeting of diaspora coordinators. “We don’t know what’s happening. We can only infer from the silence.”
The silence was itself data. A functioning city does not go silent. The erasure of communication was evidence of erasure—but evidence that proved only that something was being hidden, not what was being hidden.
The erasure rule failed against satellite observation.
The Yale Humanitarian Research Lab had been watching Darfur since the war began. Their satellites—commercial imaging platforms that anyone could access, for a price—captured what the ground-level blackouts tried to hide.
They saw the mass graves. Fresh earth disturbed in patterns consistent with burial sites. The color signature of soil changed from undisturbed to recently dug.
They saw the fires. Thermal imaging showed heat blooms across cities, patterns that matched systematic burning rather than accidental fire.
They saw the blood. In high-resolution optical imagery, the staining on concrete and asphalt was visible. The stadium in El Fasher, photographed three days after the massacre, showed dark discoloration across the field and around the trenches.
“Blood visible from space,” the researchers wrote in their report. “The scale of the killing is such that physical evidence is detectable from orbital altitude.”
The erasure rule had not anticipated satellite observation. The perpetrators were destroying evidence at ground level while leaving evidence visible from above. The gap between the rule and the countermeasure created a documentary record that no burning could eliminate.
The denial campaign was a form of erasure in narrative space.
Before the evidence could be fully compiled, the RSF was already constructing alternative accounts. The killings were defensive—they were responding to SAF attacks. The deaths were combatants—no civilians were targeted. The mass graves were for enemy soldiers—not for executed prisoners.
Social media amplified the denial. Pro-RSF accounts flooded platforms with counter-narratives. The volume was sufficient to create confusion, to make the truth one story among many, to exhaust those trying to verify what had actually happened.
Ibrahim encountered this denial when he tried to share his documentation with international journalists.
“Some people are saying the massacre didn’t happen,” a reporter told him over a satellite phone connection. “They’re saying the satellite images are fabricated, that the testimonies are SAF propaganda.”
“I was there,” Ibrahim said. “I saw the bodies. I recorded the names.”
“I believe you. But the story is contested. I need to present both sides.”
“There are not two sides. There is what happened and there are lies about what happened.”
The reporter was sympathetic but constrained by professional norms. Balance required presenting denial alongside testimony. The erasure rule operated through the conventions of journalism, exploiting the structures meant to ensure fairness.
Mass graves were the physical substrate of erasure.
A body left in the street is evidence. It can be photographed, examined, identified. A body buried in a mass grave is hidden—requiring excavation to recover, forensic expertise to analyze.
The RSF dug graves before they killed, preparing the erasure simultaneously with the execution. The wadis of Darfur contained hundreds of such graves—some documented by satellite, most unknown.
Years later, if justice ever came, the graves would need to be opened. The bodies would need to be exhumed, examined, identified through DNA or dental records or personal effects. The process would take decades. Many bodies would never be identified. Many graves would never be found.
This was the temporal dimension of erasure: pushing the evidence into a future that might never invest the resources to recover it. The rule operated not by making evidence impossible but by making evidence expensive—so expensive that it might never be extracted.
Witnesses were the most important evidence, and therefore the most important targets for erasure.
The RSF specifically hunted those who might testify: community leaders, journalists, human rights workers, anyone with the education and access to communicate what they had seen.
Ibrahim understood that his documentation made him a target. The notebooks he carried—filled with testimonies, with names, with dates and locations—were evidence that the RSF would want to destroy. And the carrier of evidence was, from the erasure perspective, just another piece of evidence.
He hid the notebooks when he crossed into Chad. He made copies and distributed them to trusted contacts. He photographed pages and uploaded them to cloud storage when he found internet access. The redundancy was protection: even if they killed him, even if they found and burned one copy, other copies would survive.
“I am already dead,” he told Khadija one night in the camp. “They will kill me if they can. I accept this. What I cannot accept is that my documentation dies with me.”
The erasure rule targeted witnesses. The documentation rule created redundancy. The competition determined what would be remembered.
The international community’s inaction was itself a form of erasure.
When atrocities are not responded to, they become normalized. When genocide is not punished, the lesson transmitted is that genocide is permissible. The failure to act erases the crime not from history but from consequence.
The UN had passed resolutions condemning the violence. The US had declared genocide. The ICC had issued arrest warrants. But the RSF commanders remained free. The gold continued flowing through Dubai. The weapons continued arriving from supply chains no one seriously disrupted.
“They’ll forget,” an RSF commander told a journalist in a rare interview, conducted under pseudonym. “The world forgets everything. In ten years, we’ll be partners again. We’ll be allies. They’ll need our gold, our cooperation, our stability. And they’ll forget what we did to get here.”
The prediction was historically grounded. The perpetrators of the first Darfur genocide—the Janjaweed commanders who had killed hundreds of thousands in 2003—had become the leaders of the RSF. The world had indeed forgotten, or at least had acted as if it had forgotten.
The erasure rule operated through memory as well as through evidence. The destruction of bodies and documents was preliminary; the real erasure was the fading of attention, the displacement by new crises, the gradual acceptance of accomplished facts.
But some evidence could not be erased.
The survivors carried memories that no burning could destroy. The diaspora maintained networks that no blackout could silence. The satellites continued watching from beyond the reach of any ground-based countermeasure.
The documentation accumulated despite the erasure efforts. Ibrahim’s notebooks. Maryam’s photographs. Leila’s databases. The Yale HRL imagery. The Amnesty International reports. The testimonies gathered by dozens of organizations across three continents.
The erasure rule was powerful but not omnipotent. It could destroy some evidence, could suppress some testimony, could create doubt and confusion. But it could not eliminate everything. The gaps in its operation allowed truth to persist.
“They want us to disappear,” Samira said at an ERR coordination meeting. “They want no one to remember. But we remember. We document. We transmit. As long as one of us survives, the truth survives.”
This was the counter-rule to erasure: preservation through redundancy, through distribution, through the sheer persistence of human memory and human networks.
Yusuf, walking west through the desert, carried evidence in his body.
The wounds he had suffered during the escape—a graze on his shoulder, cuts from the fence he had climbed through—were physical testimony to what had happened. The stadium would be cleaned, the trenches filled, the dead buried or burned. But his body remembered.
He walked for two days before finding a village that would shelter him. The villagers were Arab—normally a dangerous category given the ethnic dynamics of the war—but they were also farmers who had not joined the RSF, who retained some of the older ethics of hospitality.
They fed him. They treated his wounds. They gave him water and directions to the border.
“Why?” he asked. “I’m SAF. I’m Zaghawa. I’m everything you’re supposed to hate.”
The elder who led the village shrugged. “You’re a man who needs help. This is still our land, and our customs say we help those who need it.”
Nafeer, operating even across ethnic lines. The erasure rule could destroy evidence, but it could not destroy the primitive cultural substrate that enabled resistance.
Yusuf reached Chad on November 3, 2025. He was the highest-ranking SAF officer to survive the fall of El Fasher. His testimony, when he gave it, would be among the most detailed accounts of the siege and the massacre.
The erasure had failed in his case. One witness had escaped.
The competition between erasure and documentation would continue for years.
The immediate outcome favored erasure. Bodies were burned. Witnesses were killed. The communications blackouts prevented real-time documentation. The RSF controlled the physical space where evidence existed.
But the long-term outcome was less certain. Evidence accumulated in exile, in diaspora networks, in satellite archives, in the memories of survivors. The erasure rule had to operate continuously, forever; the documentation rule only had to succeed once, and then preserve what it had captured.
This asymmetry was fundamental. Erasure was maintenance; documentation was accumulation. Over time, the accumulation would grow while the maintenance would falter. The erasure rule required vigilance; the documentation rule required only persistence.
“They cannot burn the whole world,” Ibrahim wrote in one of his notebooks. “They can burn El Geneina. They can burn El Fasher. They cannot burn London, Cairo, Minneapolis, all the places where the truth is held. They cannot burn the satellites. They cannot burn the future.”
The statement was hope as much as analysis. But it was also structurally accurate. The erasure rule had geographic and temporal limits. Beyond those limits, the truth could persist.
The automaton continued processing. The erasure continued operating. But so did the counter-process of documentation, of memory, of witness.
The outcome was not yet determined.
Definition: The counter-rules are the inverse operations of the genocide algorithm. Where genocide identifies for elimination, counter-rules identify for assistance. Where genocide separates for killing, counter-rules connect for survival. The counter-rules do not prevent genocide; they operate in parallel, on the same population, producing survival instead of death for those they reach.
Samira had never thought of herself as running a counter-genocide.
She was running community kitchens. She was coordinating supply distribution. She was matching volunteers with needs, solving logistics problems, keeping people alive through one more day.
But as the war progressed, she began to understand her work in systemic terms. The RSF was implementing a set of procedures designed to eliminate populations. She was implementing a set of procedures designed to preserve them. The procedures were structurally parallel—identification, allocation, execution of tasks—with opposite objectives.
“We’re running the inverse algorithm,” Leila told her during one of their coordination calls. “They have rules for killing. We have rules for saving. The outputs depend on which algorithm reaches the population first.”
The framing was cold, but accurate. The war was, among other things, a race between systems. The genocide machinery and the mutual aid machinery operated simultaneously, competing for the same human substrate.
The counter-identification rule:
Where the ethnic rule identified Masalit, Fur, Zaghawa for targeting, the counter-identification rule identified anyone in need regardless of ethnicity. The criterion was not identity but situation: Who was hungry? Who was sick? Who was trapped? Who needed evacuation?
This universalism was strategic as well as ethical. A system that served only one group would be limited to that group’s capacity. A system that served everyone could draw on everyone’s capacity. The ERR networks grew because they helped everyone, and therefore everyone—Arab and non-Arab, SAF sympathizer and neutral—had reason to support them.
Aisha implemented this in Khartoum through explicit policy. “We don’t ask who you are,” she instructed new volunteers. “We ask what you need. The need is the only criterion.”
The policy created its own challenges. Some RSF fighters’ families received ERR assistance. Some people who had informed on their neighbors were fed by ERR kitchens. The universalism did not distinguish between innocent and guilty, between victim and perpetrator’s relatives.
“That’s not our job,” Aisha said when volunteers raised concerns. “Our job is to prevent starvation, not to judge who deserves to eat. Justice is someone else’s responsibility.”
The separation of functions—survival from justice—was what allowed the counter-rules to operate at scale. A system that judged would be slower, more contested, less trusted. The counter-rules operated on need alone.
The counter-separation rule:
Where genocide separated to isolate, the counter-rules connected to protect. The ERR networks were fundamentally about connection—linking those with needs to those with resources, linking the diaspora to the homeland, linking information to action.
The connection architecture was horizontal rather than hierarchical. There was no central command, no single point of control. Each node connected to adjacent nodes, and the network emerged from these connections.
This structure made the network resilient. When one node was destroyed—when an ERR coordinator was killed, when a distribution point was overrun—the network routed around the damage. Other nodes absorbed the function. The redundancy was not designed; it emerged from the connection pattern itself.
Leila visualized this in her data systems. The network topology showed clusters of dense connection, with bridges linking clusters. When a bridge was cut, the clusters found alternative paths. The system healed itself through the aggregate behavior of its nodes.
“It’s like the internet,” she told a journalist profiling the diaspora response. “Designed to survive nuclear attack. Decentralized, redundant, self-healing. Except we’re not transmitting data packets. We’re transmitting survival.”
The counter-execution rule:
This was the hardest to define, because it was not a single action but a sustained effort. The genocide’s execution rule produced death quickly. The counter-execution rule produced survival slowly, through the accumulation of small actions over time.
Every meal served was a counter-execution. Every medical treatment was a counter-execution. Every evacuation completed was a counter-execution. The rule operated not through singular events but through persistent process.
Samira measured this in her records. In the first month of war: 3,000 meals served. By the sixth month: 50,000 meals served. By the first anniversary: over 500,000 meals served. Each meal was a person kept alive for another day, a counter to the death that the genocide was trying to produce.
The mathematics was different from the execution rule’s mathematics. The genocide killed quickly and completely. The counter-rule saved slowly and partially. A person saved still needed to be saved again tomorrow. The work was never complete.
But the aggregate effect was substantial. Hundreds of thousands of people survived who would otherwise have starved. The counter-execution rule did not stop the genocide, but it reduced its scope—kept more people alive to become survivors rather than victims.
The counter-erasure rule:
Where the RSF burned and buried and denied, the ERRs documented and preserved and transmitted.
Ibrahim’s notebooks were instances of counter-erasure. Maryam’s photographs were instances of counter-erasure. Leila’s databases, compiling reports from across Sudan, were counter-erasure at scale.
The rule had two components: capture and distribute. The capture happened close to the events—witnesses writing down what they saw, photographers recording scenes, survivors telling their stories. The distribution happened across space—copies sent to multiple locations, data uploaded to cloud storage, testimonies transmitted to international organizations.
“Every copy we make is insurance,” Ibrahim told the documentation volunteers he trained in Adré. “They can destroy one copy. They can destroy ten copies. But if we make a hundred copies and send them to a hundred different places, some will survive.”
The redundancy strategy was borrowed from information theory. Claude Shannon’s work on reliable communication over unreliable channels—the same principles applied. Add redundancy. Distribute widely. Accept that some transmissions will fail, but ensure that enough succeed.
The counter-rules had limits.
They could not prevent violence—only respond to it. They could not protect people from being killed—only keep alive those who escaped killing. They operated downstream from genocide, in the space where survival was still possible.
This was the fundamental asymmetry. The genocide rules were offensive, initiating action. The counter-rules were defensive, responding to action already taken. Initiative belonged to the perpetrators; the counter-system could only react.
Samira felt this asymmetry daily. She would organize a food distribution, and then learn that the neighborhood had been attacked. The people she had planned to feed were dead or fled. The counter-rules had been too slow, the genocide rules too fast.
“We’re always behind,” she told Leila during a particularly bleak call. “We’re always responding to yesterday’s crisis while they’re creating tomorrow’s.”
“I know. But what else can we do? Arm ourselves? Become fighters?”
“Some people say we should.”
“Some people are wrong. If we become fighters, we become part of the war. We lose the neutrality that lets us operate. We lose the legitimacy that makes people trust us.”
The counter-rules required accepting their limitations. They were not a solution to genocide; they were a mitigation. The solution would require other mechanisms—international intervention, military defeat of the RSF, political transformation. The ERRs could not provide these. They could only keep people alive until the solutions arrived.
In El Fasher, during the siege, the counter-rules operated under extreme constraints.
The ERR network there was one of the most developed in Sudan—920 days of siege had forced the community to build sophisticated survival infrastructure. The counter-identification rule catalogued the needs of hundreds of thousands. The counter-connection rule linked donors and recipients across the city. The counter-execution rule delivered food and medicine despite bombardment.
Maryam had witnessed this system in operation during her year at the South Hospital. The ERR volunteers came to the hospital daily, bringing patients who needed care, taking discharged patients to recovery locations, ensuring that the medical system connected to the broader survival system.
The siege stretched everything. There was never enough food, never enough medicine, never enough volunteers. The counter-rules operated at the edge of capacity, producing just enough survival to prevent total collapse.
Then October 26 came, and the system was overwhelmed.
The RSF assault was too fast, too comprehensive. The counter-connection rule failed when communications were cut. The counter-identification rule failed when the population was suddenly reclassified—from “civilians to assist” to “targets to eliminate.” The counter-execution rule failed when the execution rule operated at industrial speed.
Many ERR volunteers were killed. They were targeted specifically because they were ERR—because they had been documented as organizers, because their work threatened the genocide’s objectives. The counter-rules did not protect their carriers.
Some counter-operations persisted even through the massacre.
In the chaos of October 26, individual acts of counter-execution continued. A nurse hiding wounded patients in a basement. A neighbor sheltering a fleeing family. A soldier who looked away at a checkpoint, allowing a truck of refugees to pass.
These were not systematic counter-rules; they were individual decisions that aligned with counter-rule logic. Each decision created survival where death was intended.
Maryam’s fourteen hours of hiding were one such decision extended over time. She had chosen to conceal herself rather than be captured. The choice kept her alive to document, to testify, to carry the counter-erasure rule forward.
Yusuf’s escape was another. The fence gap, the drainage ditch, the two-day walk—each element was a counter-execution decision, surviving against the intention of those who had captured him.
The counter-rules existed in individuals as well as in organizations. When the organizations were destroyed, the rules persisted in the humans who had internalized them.
The counter-rules could not save everyone.
This was the hardest truth. The system was imperfect, underfunded, outgunned. It saved many, but many more died despite its efforts.
In El Fasher alone, the estimated death toll in the first week after the fall was 30,000 to 50,000. The ERR network, even at its most developed, could not have saved that many against a determined assault. The counter-rules mitigated; they did not solve.
Samira struggled with this throughout the war. Every day, she made decisions about resource allocation that meant some people would be helped and others would not. The counter-identification rule could not reach everyone. The counter-execution rule could not deliver to all locations. Someone always fell through the gaps.
“How do you live with it?” a young volunteer asked her once. “How do you live with knowing that people are dying while we choose who to help?”
“I don’t know,” Samira said. “I just work. I work until I can’t work anymore, and then I sleep, and then I work again. The alternative is to stop working, and then even more people die.”
This was the existential condition of operating counter-rules: permanent inadequacy, permanent failure, permanent effort despite failure. The system worked not because it succeeded completely but because partial success was better than no effort at all.
The counter-rules were, in the end, a wager on the future.
The genocide rules aimed at permanence—the elimination of a population, the creation of accomplished facts that could not be undone. Once dead, people could not be restored to life.
The counter-rules aimed at persistence—keeping people alive until circumstances changed, preserving testimony until justice became possible, maintaining networks until reconstruction could begin.
The wager was that circumstances would change. That the war would end. That the perpetrators would fall from power. That the international community would eventually respond. That there would be a future in which survival mattered and testimony was heard.
Not everyone believed the wager would pay off. Some survivors saw only permanent darkness ahead—a future in which the genocide succeeded, the world forgot, and nothing changed.
But the counter-rules continued operating regardless of belief. Samira did not know if justice would come; she fed people anyway. Ibrahim did not know if his documentation would matter; he wrote anyway. Leila did not know if her databases would ever be used in prosecution; she built them anyway.
The counter-rules were faith made operational. Faith that the future was worth preserving people for. Faith that the truth was worth documenting. Faith that survival was meaningful even when its meaning was not yet visible.
The automaton processed genocide. The counter-automaton processed survival. Both continued operating, in parallel, on the same population, producing different outputs.
The outcome would be determined by which system prevailed.
Definition: Rule collision occurs when opposing systems attempt to operate on the same substrate simultaneously. When genocide rules and counter-rules meet at the same location, at the same time, concerning the same population, the outcome is determined by relative power, timing, and contingency. Collision is where the abstract logic of rules meets the concrete chaos of implementation.
The collision was most visible at checkpoints.
The RSF established checkpoints to control movement—implementing the ethnic rule, filtering populations, directing targets toward execution sites. The ERR networks developed methods to bypass checkpoints—implementing counter-connection, maintaining supply lines, evacuating vulnerable people.
A truck approaches a checkpoint. Inside: a family, classified as target ethnicity. Also inside: supplies for an ERR kitchen. Also inside: a driver who knows the checkpoint commander, who has paid bribes, who has developed a reputation for transporting goods rather than people.
The collision happens in the checkpoint interaction. The RSF’s ethnic rule says: inspect the truck, identify the family, apply the separation procedure. The ERR’s counter-connection rule says: move the family through, use the driver’s relationship, exploit the gap in the system.
Outcome: indeterminate. It depends on which guard is on duty, whether the bribe is sufficient, whether the family is discovered, whether the checkpoint commander is feeling generous or suspicious that day.
The rules provide logic, not certainty. The collision introduces variance.
Aisha had developed checkpoint bypass protocols by August 2023.
The protocols were based on accumulated experience—what worked, what failed, what patterns could be exploited. They were transmitted through the ERR network as informal knowledge, updated as conditions changed.
Use regular drivers. A driver who passed the same checkpoint daily, transporting known goods, aroused less suspicion than a new face. The counter-connection rule could exploit the ethnic rule’s reliance on unfamiliarity.
Document cargo openly. Supplies for hospitals, for kitchens, for obvious humanitarian purposes were sometimes waved through. The documentation provided cover—even if the truck also carried passengers hidden beneath the supplies.
Time the passage. Checkpoint vigilance varied. Early morning, when guards were tired. Prayer times, when attention was elsewhere. The counter-rules could exploit the execution rule’s temporal patterns.
Accept losses. Some convoys would be stopped, inspected, stripped. The counter-rules could not save everyone. The goal was to move enough through that the network continued functioning.
The protocols were not foolproof. Convoys were captured, drivers killed, passengers discovered and executed. The collision was always uncertain. But the protocols increased the probability of survival—shifted the odds slightly toward counter-rule success.
In El Fasher, the collision was prolonged over 920 days.
The siege was a sustained interaction between genocide rules and counter-rules, each side adapting to the other’s operations.
The RSF’s siege rule created scarcity. The ERR’s counter-execution rule developed urban farming, rooftop gardens, improvised agriculture that produced some food despite the blockade.
The RSF targeted water infrastructure. The ERR developed well-digging teams, identified hidden water sources, created purification systems from available materials.
The RSF bombarded during food distributions. The ERR varied distribution times and locations, never using the same site twice, treating predictability as vulnerability.
The collision produced innovation on both sides. The genocide rules adapted to counter-rule tactics. The counter-rules adapted to genocide adaptations. The system co-evolved, each side learning from the other’s moves.
Colonel Yusuf observed this co-evolution from the military perspective. “It’s like a game,” he told Ismail. “But the pieces are people, and losing a piece means dying.”
The collision created moral complexity.
Consider: an ERR volunteer learns that an RSF checkpoint commander’s daughter is sick. The counter-identification rule says: she needs help. The practical calculus says: helping her might create goodwill, might make the commander less vigilant, might allow more refugees to pass.
Does the volunteer help? Is helping the enemy’s child collaboration or strategy? Is refusing to help a child because of her father’s crimes justice or cruelty?
Samira faced this in October 2023, when an RSF family requested assistance from an ERR kitchen in Khartoum.
“They’re RSF,” a volunteer said. “Their father kills our people. Why should we feed them?”
“They’re children,” Samira said. “They didn’t choose their father.”
“Every meal we give them is a meal we can’t give to our people.”
“Our people includes everyone. That’s the principle.”
The decision was to provide assistance. The principle of universalism was maintained. The volunteer who objected left the network.
Later, Samira wondered if she had been right. The principle was clear, but principles operated in a context where resources were finite and enemies were real. The collision of rules created situations where no choice was clean.
The collision sometimes produced cooperation.
Not between the RSF as an organization and the ERRs, but between individuals whose positions in the systems did not perfectly align with their behaviors.
An RSF soldier who warned a neighborhood before an operation—breaking the execution rule to honor some older loyalty or residual humanity.
A SAF officer who allowed ERR supplies through military checkpoints—breaking military protocol to support civilian survival.
An Arab tribal leader who sheltered Masalit refugees—breaking the ethnic solidarity that the RSF expected, honoring the older ethics of nafeer.
These individual deviations did not change the systems, but they created gaps through which some people escaped. The collision at the individual level produced different outcomes than the collision at the systemic level.
Yusuf’s rescue by Arab farmers was such a gap. The ethnic rule said: he is Zaghawa, an enemy. The nafeer rule said: he is a man in need. The farmers chose nafeer over ethnicity. Yusuf survived because of their choice.
The collision in digital space was as intense as the collision in physical space.
The RSF’s erasure rule targeted communication infrastructure—destroying cell towers, blocking internet access, creating information blackouts. The counter-erasure rule developed alternatives—satellite phones, mesh networks, USB drives carried by trusted couriers.
Leila coordinated this information war from London. When the main communication channels failed, she helped activate backup channels. When the backup channels were discovered and targeted, she helped develop new ones.
“It’s an arms race,” she told the diaspora coordination team. “They block; we route around. They jam; we find new frequencies. They can’t win completely because technology keeps evolving. We can’t win completely because they have more resources.”
The collision in information space determined what the world knew about what was happening. When the genocide rules succeeded, silence resulted. When the counter-rules succeeded, documentation emerged.
The El Fasher blackout of October 2025 was a genocide-rule victory in information space. For three days, almost nothing emerged from the city. The massacre happened in documentary darkness.
Then the counter-rules reasserted themselves. Survivors reached Chad with testimonies. Satellite imagery showed the aftermath. The erasure was incomplete. The collision continued.
The collision could flip suddenly.
A location that was counter-rule territory could become genocide territory in hours. A location that was genocide territory could be abandoned, allowing counter-rules to operate.
In Khartoum, the control of neighborhoods shifted repeatedly during 2023 and 2024. An area would be under RSF control, the genocide rules operating, the population subject to ethnic targeting. Then SAF would retake the area, the genocide rules would withdraw, the counter-rules would resume.
The oscillation created its own trauma. People never knew which system they would wake up under. The rules that determined their survival could change overnight.
Maryam had worked through these oscillations. One week, the hospital was in RSF territory, serving whoever came but knowing that ERR operations were suppressed. The next week, SAF controlled the area, ERR volunteers returned, the hospital became a node in the counter-network again.
“The rules are the same,” she observed. “But which rules are active depends on who holds the ground. The collision is decided by military force, and then the systems operate accordingly.”
This was the relationship between war and genocide: war determined territory, and territory determined which rules applied. The ERRs could not operate where the RSF held total control. The genocide could not operate where the SAF held.
Some collisions were deliberately engineered.
The ERRs developed a practice of “embedded survival”—placing counter-rule operations inside genocide territory so that they would be present when collisions occurred.
A safe house in an RSF-controlled neighborhood. A medical cache hidden near a detention site. A documentation station positioned to record what happened when massacres occurred.
The logic was to ensure that counter-rules were present at collision points, ready to operate when gaps appeared. The genocide could not be perfectly sealed; there were always moments of chaos, confusion, opportunity. The counter-rules waited for these moments.
Ibrahim’s documentation in El Geneina was an example. He had remained in the city as the genocide rules closed in, recording testimonies until the last possible moment. When he fled, he carried the documentation with him—counter-erasure materials extracted from the collision zone.
The strategy was dangerous. Many who embedded themselves in genocide territory did not survive. But some did, and they carried evidence and testimony that would not have existed otherwise.
The collision at El Fasher on October 26, 2025 was the largest of the war.
All the rules operated simultaneously, at maximum intensity, on a population of hundreds of thousands.
The siege rule reached its terminus. The ethnic rule activated. The separation rule operated in the stadium and across the city. The execution rule processed thousands. The erasure rule burned buildings and cut communications.
The counter-rules operated too. ERR volunteers hid refugees. Medical staff treated the wounded despite the danger. Documentation continued—Maryam’s photographs, others’ testimonies, evidence preserved against the erasure.
The collision lasted approximately 72 hours, until the genocide rules achieved dominance. After that, counter-rules could operate only in hiding, only for those already concealed, only in the gaps that remained.
The outcome was catastrophic. Tens of thousands died. The ERR network in El Fasher was destroyed. The counter-rules were defeated in that location, at that time.
But not eliminated entirely. Survivors fled with their knowledge. Documentation emerged despite the erasure. The counter-rules persisted in exile, ready to reconstitute when conditions permitted.
The collision would continue beyond any individual battle.
The war was not a single collision but a continuous series of collisions across space and time. Genocide rules advanced in one location while counter-rules operated in another. Territory changed hands. Populations moved. The collision zone shifted.
This was the nature of the conflict: not a decisive battle but an ongoing competition between systems. The genocide aimed at completion—the total elimination of target populations. The counter-system aimed at persistence—keeping enough people alive, preserving enough evidence, maintaining enough capacity to reconstitute.
Neither system could achieve total victory. The genocide could not kill everyone; survivors always escaped. The counter-rules could not save everyone; many died despite every effort.
The collision would be resolved, eventually, by forces outside both systems. International intervention, if it came. Military exhaustion, if resources depleted. Political change, if power structures shifted. The collision itself would not determine the outcome; the collision was the mechanism through which the outcome would be produced.
Part II was complete. The transition rules—adjacency, ethnicity, siege, separation, execution, erasure, counter-rules, collision—were established.
The automaton was running. The rules were operating. The collision was ongoing.
Part III would examine propagation: how the patterns spread.
Definition: In cellular automata, Generation Zero is the initial configuration—the pattern that seeds all subsequent evolution. The El Geneina massacre of June 2023 was the genocide’s Generation Zero. The pattern established there would replicate across Darfur and beyond. Understanding Generation Zero is essential to understanding everything that followed.
The city had been preparing for war without knowing it.
El Geneina, capital of West Darfur State, population approximately 500,000 before the displacement crisis. A city of markets and mosques, of schools and hospitals, of ordinary people living ordinary lives. A city that had survived the first Darfur genocide in 2003 and had been rebuilt, partially, through the intervening years.
The Masalit were the majority population—the ethnic group that had lived in this region for centuries, that had resisted incorporation into Sudan’s Arab-dominated state, that had maintained their language and customs despite pressure to assimilate.
The Arabs were the minority—traders, herders, some who had settled permanently, others who passed through seasonally. The relations between groups were complex: intermarriage existed, trade was constant, conflict was recurrent but not continuous.
When the war began in Khartoum on April 15, 2023, El Geneina’s residents watched with apprehension. The fighting was far away, but the RSF had strength in Darfur. Everyone remembered 2003.
“It will reach us,” Khadija told Ibrahim on the first day. “It always reaches us.”
She was right, but the timing was uncertain. The pattern had not yet propagated.
The fighting arrived on April 24, 2023.
The RSF attacked the SAF garrison in El Geneina as part of their broader offensive across Darfur. The initial fighting was military: RSF versus SAF, armed forces against armed forces. The civilian population sheltered and waited.
But the military fighting was prelude, not main event. The RSF was not primarily interested in defeating the SAF garrison. They were interested in something else—something that the military operations were designed to enable.
Over the following weeks, as the SAF garrison weakened, the RSF’s operations shifted. The ethnic rule activated. Masalit neighborhoods came under attack—not as collateral damage from combat, but as deliberate targets.
Ibrahim watched the pattern emerge. The checkpoints that sorted people by ethnicity. The young men taken away and not returned. The houses burned in specific neighborhoods while adjacent Arab houses stood untouched.
“It’s 2003 again,” Khadija said. “The same pattern. They’re doing the same thing.”
Ibrahim documented. He recorded testimonies from those who escaped the early attacks. He noted locations, described perpetrators, estimated casualty numbers. The teacher in him insisted on precision; the witness in him insisted on preservation.
June 2023 was the peak.
The SAF garrison surrendered or withdrew—the accounts varied—on June 9. What followed was unrestrained: the siege rules, the separation rules, the execution rules, all operating without military opposition.
Governor Khamis Abakar was captured and killed on June 15. His death was filmed and distributed—not an erasure but a display, a message to other leaders who might consider resistance or documentation.
The killing accelerated. Mass executions in the wadis outside the city. Door-to-door operations in Masalit neighborhoods. Systematic rape as a weapon of terror. The pattern that would replicate in subsequent massacres was established here, in these June days.
The death toll was estimated at 10,000 to 15,000 in El Geneina alone. The precision was impossible—too many bodies burned, too many buried in unknown locations, too many washed away in the seasonal rains. But the order of magnitude was clear.
Generation Zero was complete. The pattern was established.
Ibrahim fled on June 9, before the worst began.
He and Khadija joined the stream of refugees moving west, toward Chad. The route was dangerous—RSF forces patrolled the roads, hunting those who fled. But the alternative was remaining in a city that had become a killing ground.
They traveled with others, a group that grew and shrank as people joined and separated. Safety in numbers, or at least in dispersed targets. The RSF could not pursue everyone; some would escape.
The journey took five days. They walked mostly at night, hiding during daylight. They passed through villages in various states—some still inhabited, some abandoned, some destroyed. The pattern was propagating even as they fled from it.
On the third day, they encountered a massacre in progress.
A village—Ibrahim could not remember the name afterward, or perhaps he had never learned it—was burning. RSF forces were visible, moving through the streets. Bodies lay in the open. The refugees hid in a wadi, barely breathing, until the forces moved on.
“We can’t help them,” Khadija whispered. “We can only survive.”
Ibrahim knew she was right. The counter-rules had no capacity here; survival was the only available response. But the knowledge felt like failure. He was fleeing while others died.
The pattern that Generation Zero established had specific features:
Speed. The killing was rapid—compressed into days rather than weeks. The RSF had learned from 2003 that prolonged operations attracted international attention. Speed reduced the window for intervention.
Totality. The targeting was comprehensive—all Masalit, not selected individuals. The ethnic rule operated without exceptions, without the pretense of distinguishing combatants from civilians.
Documentation. The perpetrators documented their own crimes—videos of killings, photographs of bodies, social media posts celebrating the destruction. The erasure rule would operate later, but in the moment, the perpetrators wanted their work known.
Coordination. Multiple units operated simultaneously, covering the city systematically. The operation was planned, not spontaneous—the cellular automaton executed according to a designed pattern.
Impunity. The killing occurred in full view of the international community, which did nothing. The genocide rules had been tested against the possibility of intervention and found that no intervention would come.
These features—speed, totality, documentation, coordination, impunity—would characterize subsequent massacres. Generation Zero was not merely the first instance; it was the template that subsequent instances would copy.
The survivors of El Geneina carried the pattern with them.
Those who reached Chad brought testimonies, photographs, memories. They described what they had seen: the methods, the perpetrators, the sequence of operations. The documentation accumulated in the camps, creating a record of Generation Zero that would persist beyond the erasure efforts.
Ibrahim’s notebooks were part of this record. In Adré, he continued documenting, collecting testimonies from other survivors, building a comprehensive account of what had happened.
“Why do you do this?” a fellow refugee asked him one night. “Why write down the horrors? Isn’t it enough that we survived?”
“Because if no one writes it down, they win,” Ibrahim said. “They can say it didn’t happen. They can deny. But if we have the record—the names, the dates, the descriptions—then the truth exists. And maybe someday, the truth will matter.”
The fellow refugee was skeptical. Truth had not mattered in 2003. Truth had not prevented the RSF from becoming a legitimate force. Truth seemed powerless against men with guns.
But Ibrahim continued writing. The counter-erasure rule operated through him, preserving what the perpetrators wanted destroyed.
Generation Zero propagated through multiple channels.
The perpetrators learned from El Geneina. They refined their methods, identified what worked, prepared to replicate the pattern in other locations. The genocide algorithm was updated with operational experience.
The command structure transmitted lessons. Units that had participated in El Geneina were deployed to other areas, carrying their knowledge with them. The carriers of genocide primitives spread across Darfur.
The documentation spread too—but selectively. The perpetrators’ videos circulated on certain platforms, serving as both celebration and threat. The survivors’ testimonies circulated on other platforms, serving as evidence and alarm.
The international community observed Generation Zero and calibrated its response: condemnation without consequence. This non-response was itself a propagation—a signal that the pattern could be replicated without triggering intervention.
Leila watched Generation Zero from London with growing horror.
Her coordination platform showed the data streams from El Geneina: first the warnings, then the fragmentary reports during the massacre, then the flood of survivor testimonies afterward. The pattern was visible in the data—the spike in violence, the communications blackout, the subsequent documentation.
“We knew it was happening,” she said later. “We could see it in real time. But we couldn’t do anything. We could only watch.”
The watching was not nothing. Leila’s platform aggregated the information, made it visible, transmitted it to journalists and human rights organizations. The documentation that Ibrahim and others created in the camps was amplified by the diaspora network.
But watching was not intervention. The counter-rules could document; they could not prevent. Generation Zero occurred while the world watched, and the world’s watching did not stop it.
“They’re going to do it again,” Leila told her team after the first reports from Ardamata emerged in November. “They’re following the same pattern. Generation One is starting.”
The prediction was correct. The pattern was propagating.
Ardamata, November 4, 2023: Generation One.
The pattern repeated with high fidelity. A neighborhood attacked. Men separated and executed. Women subjected to systematic violence. Buildings burned. Erasure operations targeting evidence and witnesses.
The death toll: 800 to 1,300 in a single day. Smaller than El Geneina but identical in structure. The cellular automaton had iterated, producing a new generation from the template of the previous one.
The replication fidelity was high—the methods, the targeting, the sequence of operations matched Generation Zero closely. The perpetrators had learned the pattern and were reproducing it.
Other locations followed. Nyala, Zalingei, villages across Darfur whose names would not reach international media but whose destruction followed the same template.
The pattern was spreading like a wavefront, emanating from the point of origin, transforming each location it touched.
Generation Zero was the seed.
In the mathematics of cellular automata, the initial configuration determines everything that follows. The rules operate on the configuration, producing new configurations, but the rules cannot create something from nothing. They need a starting pattern.
El Geneina was that starting pattern. The massacre there established the template—the specific combination of rules, the operational methods, the targeting criteria. Every subsequent massacre was a transformation of this initial configuration.
The seed contained the genocide’s DNA: the ethnic ideology, the execution protocols, the erasure methods. The DNA replicated as the pattern spread. The variations between massacres were minor; the core was constant.
Understanding Generation Zero was therefore essential to understanding the entire genocide. The pattern could not be predicted from first principles; it had to be observed in its initial manifestation. Once observed, it could be recognized wherever it appeared.
Ibrahim understood this, intuitively if not formally. His documentation of El Geneina was not merely a record of one massacre. It was a record of the template that would repeat across Sudan. By documenting Generation Zero, he was documenting the algorithm itself.
The automaton continued to iterate. Generation One led to Generation Two. The wavefront spread. But the pattern remained recognizable—the same operations, the same targeting, the same sequence.
El Geneina had seeded the genocide. The harvest was still being collected.
Definition: Within genocide systems, articulators are high-priority targets—individuals capable of naming what is happening, of communicating to external audiences, of organizing resistance. The articulator targeting rule operates within the broader ethnic rule but with additional intensity. Governor Khamis Abakar was the paradigmatic articulator of El Geneina. His death demonstrated how the system processes those who can speak.
Khamis Abakar was, in June 2023, the highest-ranking Masalit official in Sudan.
As governor of West Darfur State, he had navigated impossible terrain for years: representing his people within a state structure that was often hostile to them, maintaining relationships with Arab tribal leaders while defending Masalit interests, seeking peace in a region where peace was always provisional.
He was fifty-three years old, educated, articulate in Arabic and English, connected to national and international networks. He had met with UN officials, with ambassadors, with journalists. He understood how to communicate beyond the borders of his state.
When the war began, he attempted what he had always attempted: negotiation. He reached out to RSF commanders, seeking to prevent the fighting in Khartoum from spreading to El Geneina. He offered agreements, accommodations, whatever might preserve his city.
The negotiations failed. The RSF was not interested in accommodation; they were interested in elimination.
As the violence escalated in June, Abakar’s role shifted from negotiator to witness.
He could see what was happening. He received reports from across the city—massacres in this neighborhood, mass graves in that wadi, systematic targeting of Masalit civilians. He understood that this was not war but genocide.
On June 15, he spoke to Al Jazeera Arabic. The interview was conducted by phone—the city was too dangerous for journalists to enter. His voice was the voice of El Geneina to the outside world.
“What is happening is not war,” he said. “It is ethnic cleansing. They are killing people based on their ethnicity. This is genocide.”
The words were precise. He named what he saw: ethnic cleansing, genocide. The terms had legal meaning, carried weight under international law, demanded response.
He knew the risk. By naming the genocide, he was marking himself as the most dangerous articulator in the region. The erasure rule demanded his elimination.
Hours after the interview, the RSF captured him.
The details of the capture were disputed afterward. Some accounts said he was taken at a checkpoint, attempting to flee. Others said he was extracted from a safe house, betrayed by someone who knew his location. The RSF’s own account claimed he was a combatant captured in battle.
What was not disputed: by June 16, Khamis Abakar was dead.
The manner of death was also disputed. Some witnesses said he was executed immediately, shot in the back of the head in the style of military execution. Others said he was tortured first, his death prolonged as punishment for his articulation. The RSF denied killing him, claimed he died “in custody” from unspecified causes.
Videos emerged—as videos always emerged in this war, documentation serving both erasure and evidence. A video showed a body that survivors identified as Abakar, stripped, bearing signs of violence. The video was distributed on social media by RSF supporters, a trophy of their success in eliminating the articulator.
The targeting of Abakar was not exceptional; it was exemplary.
The genocide system prioritized articulators because articulators were force multipliers. One governor could communicate to millions. One interview could shape international perception. One witness with access to media could create pressure for intervention.
The execution of Abakar served multiple purposes:
Elimination. He could no longer speak, no longer organize, no longer represent his people to the world.
Warning. Other potential articulators saw what happened to those who named the genocide. The warning suppressed speech.
Demonstration. The RSF showed that it could reach anyone, including the highest-ranking officials. No one was safe.
The articulator targeting rule operated throughout the genocide. Teachers were killed because they could educate. Journalists were killed because they could report. Community leaders were killed because they could organize. Doctors were killed because they understood injury patterns and could testify to violence.
Ibrahim, as a teacher who documented testimonies, understood that he was in this category. His survival was partly luck, partly the chaos that created gaps, partly his decision to flee before the system processed him.
The international community’s response to Abakar’s death was indicative of the broader response to the genocide.
Condemnation was issued. The UN expressed concern. Human rights organizations documented the killing. Journalists wrote stories. The words accumulated.
But no action followed the words.
Abakar had called for intervention. He had said, in his final interview, that the international community must act to prevent genocide. The call went unanswered. He was killed. The genocide continued.
The lesson transmitted by the non-response was clear: articulation did not produce protection. Speaking did not save you. The international community would witness but would not intervene.
This lesson shaped subsequent behavior. Potential articulators understood that speaking carried risk without corresponding benefit. Some spoke anyway—the imperative of witness overriding calculation. Others fell silent, survival requiring invisibility.
Ibrahim thought of Abakar often in the months that followed.
The governor had been his superior in a sense—the representative of the state in which Ibrahim was a citizen, the leader of the Masalit community to which Ibrahim belonged. They had never met personally, but Ibrahim had known of him, had respected him, had hoped that his negotiations might succeed.
The death was personal in a way that the anonymous masses of victims were not. A named individual, with a history, with a family, with a face that could be pictured. The death of ten thousand was a statistic; the death of Khamis Abakar was a tragedy.
“He spoke for us,” Ibrahim told Khadija in the camp. “He said what we all knew, to the whole world. And they killed him for it.”
“Would you have done the same?”
“I don’t know. He was braver than me. Or more foolish. Or perhaps there is no difference.”
Ibrahim continued documenting, continued being an articulator in his own smaller way. His notebooks would never reach Al Jazeera. But they might reach historians, might reach courts, might reach the future that Abakar had tried to create through his speech.
The articulator targeting rule created a documentation paradox.
The system wanted to eliminate those who could speak. But eliminating articulators created evidence—the deaths themselves were documentable, the pattern of targeting revealed the system’s logic.
Abakar’s death, precisely because he was known, became internationally visible. His name was invoked in UN sessions, in human rights reports, in calls for justice. The attempt to silence him amplified his voice posthumously.
This was the paradox of erasure: the most complete erasure would be invisible, but invisible erasure was operationally impossible. Every act of elimination left traces. Every articulator killed became evidence of articulator targeting.
The system could never achieve perfect silence. It could only shift the ratio—more people killed than testified, more evidence destroyed than preserved. The competition between erasure and documentation was asymptotic, never reaching either zero or infinity.
Other articulators followed Abakar.
Journalists who remained in the conflict zones, reporting despite the danger. Human rights workers who documented despite the targeting. Community leaders who organized despite the risks.
Some survived. Many did not. The cumulative death toll among articulators was disproportionate—far exceeding their share of the population, reflecting the system’s prioritization.
The survival of articulators depended on the counter-rules’ capacity to protect them. ERR networks helped evacuate journalists. Diaspora connections helped sustain communication links. Documentation networks maintained redundancy so that the loss of any individual articulator did not mean the loss of their testimony.
“We are all Abakar now,” Samira said at an ERR coordination meeting. “Anyone who documents, anyone who speaks, anyone who organizes—we are all targets the way he was a target. We just don’t have his profile.”
The observation was not paranoid but accurate. The articulator targeting rule operated continuously, seeking those who could name what was happening.
Abakar’s family survived.
His wife and children had been evacuated before the final collapse, through channels that the governor had maintained precisely because he understood the risk. They reached Chad, then Cairo, then further exile.
In Cairo, his wife spoke to journalists, continuing the articulation that her husband had begun. She described his final days, his attempts to save the city, his decision to speak despite the danger.
“He knew they would kill him,” she said. “He spoke anyway. He said that silence was death too—that if no one named what was happening, it would be as if it never happened.”
The articulation passed through generations. Abakar’s children, growing up in exile, would carry their father’s legacy. They would know what he had done and why he had died. They would become, in their own ways, articulators of the memory he had died to preserve.
This was the temporal dimension of the articulator rule: killing an individual did not kill their influence. The dead could still speak through those who remembered them.
Governor Khamis Abakar was buried in an unmarked grave somewhere in the outskirts of El Geneina.
The location was unknown. The RSF disposed of the body in the manner they disposed of thousands of others—quickly, without ceremony, in a place calculated to prevent subsequent recovery.
But his words survived. The Al Jazeera interview was archived, copied, preserved in the distributed memory of the internet. His final articulation—”This is genocide”—became a defining statement of the El Geneina massacre.
The erasure rule had eliminated the man. The counter-erasure rule had preserved his message. The collision between the two rules produced a partial outcome: death and memory, silence and echo.
He had spoken. He had died. But the speech persisted beyond the death.
This was what articulation could achieve, even when the articulator was eliminated. The words outlasted the body. The testimony outlasted the witness. The naming of genocide outlasted the attempt to erase both the genocide and the name.
Abakar’s ghost would haunt the subsequent proceedings—the trials, if trials came; the histories, when histories were written; the memory, as long as memory persisted.
The articulator was dead. The articulation continued.
Definition: The road in genocide systems is not merely geography but a processing zone. Populations attempting to flee become linearly concentrated along limited routes, creating vulnerability. The road is where the execution rule can operate efficiently against moving targets—where escape itself becomes a trap.
The distance from El Geneina to the Chadian border is approximately seventy kilometers.
In a vehicle, on a paved road, in peacetime: an hour’s drive. On foot, at refugee pace, carrying children and elderly, dodging checkpoints and patrols: five to seven days. The distance was not merely physical; it was a gauntlet.
The RSF understood that flight was part of the genocide. A population that fled could be pursued, intercepted, processed along the route. The roads to Chad became killing corridors—places where the ethnic rule operated against people already classified as targets by their decision to flee.
Ibrahim and Khadija traveled this route in June 2023, one group among thousands making the same desperate journey. They carried water, food, the notebooks that Ibrahim had filled with testimonies. They carried also the knowledge that the road was death as much as it was escape.
The first checkpoint was fifteen kilometers outside El Geneina.
The RSF had positioned themselves at a junction where two routes merged—a natural collection point for refugees. Their procedure was established: stop the column, inspect, separate men from women, process accordingly.
Ibrahim’s group saw the checkpoint from a distance. They were traveling at night, but the fires and vehicle lights were visible. Someone in the group—a young man who knew the terrain—led them into a wadi, a detour that added three kilometers but avoided the checkpoint.
“There’s another one at Ardamata,” the young man said. “And another near the border. We have to go through the bush, not the road.”
The detour strategy worked for Ibrahim’s group. They passed the checkpoint zone without being stopped. But the strategy required knowledge, mobility, luck. Many groups did not have these resources.
The killing on the roads was documented by those who survived.
Halima, who had escaped El Geneina on June 12, described what she saw on the route to Chad:
“The first bodies were at kilometer twenty. A group that had been stopped, we could see. The men were separate from the women, lined up. All dead. The women were still there, sitting by the road. Some were injured. They joined our group, those who could walk.”
The bodies accumulated as the distance increased. Each checkpoint zone produced its output. The refugees walking to Chad walked past the evidence of what happened to those who had not evaded the checkpoints.
“By kilometer forty, we stopped counting,” Halima said. “There were too many. We walked past them. We couldn’t help them. We could only keep walking.”
The road became a timeline of the genocide—the freshest bodies near the city, the older bodies decomposing further toward the border. The pattern was legible to anyone who traveled the route.
The RSF’s operations on the roads were not improvised.
The checkpoints were positioned strategically, at places where geography concentrated movement. The patrols swept the bush on either side of the roads, hunting those who tried to detour. The coordination suggested planning, not opportunistic violence.
The purpose was not merely killing. It was also looting—the refugees carried whatever wealth they had been able to bring, and that wealth could be extracted. And it was also terror—the violence on the roads sent a message to those still in the city, discouraging flight, trapping populations for subsequent processing.
The road violence was an extension of the city violence, not separate from it. The genocide algorithm operated across the entire space between El Geneina and Chad, treating the road as a processing zone equal to the neighborhoods.
Ibrahim’s group took six days to reach the border.
They walked at night, hid during the day. They ate what they had brought, rationed carefully. They drank from the seasonal pools that dotted the landscape, knowing the water might be contaminated but having no alternative.
The group grew and shrank. People joined from other routes, from villages along the way, from the bush where they had been hiding. People left—some too weak to continue, some choosing different routes, some simply disappearing.
On the fourth night, they heard gunfire behind them. A pursuing patrol, perhaps, or another checkpoint encounter. They moved faster, the exhausted carrying the more exhausted, covering eight kilometers in darkness that should have taken four.
On the fifth day, hiding in a dry riverbed, they were found.
The patrol was four men on two motorcycles.
They had seen the group’s tracks and followed them into the riverbed. They were RSF—the uniforms, the weapons, the Arabic commands. They ordered everyone to stand, to line up.
Ibrahim expected the separation to begin. Men on one side, women on the other. The familiar procedure.
But something was different. One of the patrol members—the youngest, perhaps twenty—was speaking rapidly to the leader. Ibrahim could not hear the words, but he could see the gestures. Pointing at the group, shaking his head, arguing.
The leader listened. He looked at the group—forty-seven people, women and children outnumbering men, all of them clearly refugees, clearly harmless, clearly exhausted.
He gave an order. The patrol members lowered their weapons. The leader spoke to the group in Arabic:
“Go. Quickly. Before others come.”
They went. No one asked why. No one delayed to understand. They fled the riverbed and did not stop for hours.
Later, in the camp, Ibrahim would think about that moment. A gap in the system. An individual deviation from the algorithm. One patrol leader who, for whatever reason, chose not to execute the rule.
The group survived because of that choice. The automaton was not omnipotent; its carriers were human, and humans sometimes failed to carry.
The border crossing was itself a checkpoint, but a different kind.
The Chadian military had established a presence at the main crossing points, attempting to manage the influx. Their procedure was not killing but processing—registration, health screening, direction to camps.
The relief of crossing into Chad was physical. Ibrahim felt it in his body—a loosening of tension he had not known he was carrying. For six days, every moment had been danger. Now, suddenly, the danger was behind a line on a map.
“We made it,” Khadija said. It was the first complete sentence she had spoken in two days.
“We made it,” Ibrahim agreed. But even as he said it, he thought of those who had not. The bodies on the road. The groups stopped at checkpoints. The people still in El Geneina, still subject to the processing.
They had escaped the genocide. The genocide continued.
The scale of the exodus was unprecedented.
In June 2023 alone, approximately 250,000 people crossed from West Darfur into Chad. They arrived at border points overwhelmed by the numbers, at camps that had not existed a month before, at a humanitarian system scrambling to respond.
The roads to Chad had become one of the largest forced migrations in the world. The processing that the RSF intended—the killing of target populations—had produced an output the RSF had not fully intended: a survivor population, concentrated in Chad, documented and witnessed.
The road violence reduced the survivor population but could not eliminate it. The execution rule operated against those it caught, but the counter-movement of flight produced escapees. The collision between genocide and exodus determined who lived.
Leila tracked the exodus in her data.
The diaspora coordination platform showed the flow: reports from the border crossings, tallies of arrivals at camps, need assessments from aid organizations. The data visualized the exodus as a wave—thousands moving westward, then tens of thousands, then hundreds of thousands.
“They’re emptying Darfur,” she told the coordination meeting. “The population that survived the killing is being pushed into Chad. We’re seeing the largest displacement since the original Darfur genocide.”
The data also showed what the exodus left behind: silence. The areas that had been evacuated stopped producing data. No ERR reports, no survivor testimonies, no communication of any kind. The erasure was complete in those zones.
“What’s happening there now?” someone asked.
“I don’t know. No one knows. Everyone who could tell us is either dead or in Chad.”
The road to Chad was the last data pipeline from the genocide zones. Those who survived the road brought information. Those who died on the road took their knowledge with them.
Ibrahim and Khadija arrived at Adré camp on June 14, 2023.
They were processed, registered, assigned a space in a section designated for new arrivals. The space was a patch of ground with a plastic sheet for shelter. It was inadequate, undignified, barely survivable.
It was also safety.
Ibrahim began documenting again almost immediately. The camp filled with survivors, each carrying testimonies. The road had brought them here, and now their stories could be collected, preserved, transmitted.
“We are the evidence,” he told a group of survivors he was interviewing. “We walked through the genocide and lived. Everything we saw, everything we remember, is evidence. They cannot erase us.”
The statement was aspirational as much as factual. The RSF could not reach them in Chad, but the erasure continued in Sudan. The road had saved those who traveled it, but the road could not save those who never reached it.
The camp grew daily. The road kept delivering survivors. The documentation continued.
The genocide was ongoing, but so was the counter-process of witness.
Definition: Generation One in cellular automata is the first iteration after the initial configuration—the pattern produced when the rules operate on Generation Zero. Ardamata, November 4, 2023, was Generation One of the Sudan genocide. The El Geneina template replicated with high fidelity, demonstrating that the pattern was not a singular event but a reproducible procedure.
Ardamata was a neighborhood on the outskirts of El Geneina.
It had survived June’s massacre partially—its distance from the city center, its different ethnic composition, the chaos of the initial operations. Many residents had fled, but some remained. The neighborhood existed in a liminal state: not safe, not yet destroyed.
Through the summer and fall of 2023, Ardamata became a refuge for those displaced from elsewhere in El Geneina. The Masalit population concentrated there, hoping that the violence had passed, that the RSF would be satisfied with what they had already destroyed.
The hope was miscalculated. The genocide algorithm was not satisfied; it was iterating.
The operation began on November 4, 2023.
The RSF surrounded Ardamata before dawn, establishing a perimeter that prevented escape. They had learned from June: some populations had escaped because encirclement was incomplete. This time, the encirclement was thorough.
At first light, the units moved in. The procedure was familiar—the same procedure that had been executed in central El Geneina five months earlier.
Separation first. Men from women. Fighting-age males identified and collected. The separation happened at multiple points simultaneously, the coordination tight, the operation systematic.
Then execution. The men were taken to sites outside the neighborhood—a school, a field, locations that could accommodate the processing. The killings began mid-morning and continued through the day.
The death toll: 800 to 1,300 in approximately eight hours. The estimate range reflected the usual uncertainty—bodies burned, bodies buried, bodies uncounted.
The replication fidelity was high.
Comparing El Geneina (June) to Ardamata (November), the pattern matched closely:
Encirclement before assault. Both operations began with perimeter establishment, trapping the population.
Separation by gender and age. The same sorting criteria applied in both cases.
Execution at designated sites. The wadis and fields outside El Geneina, the school and fields outside Ardamata—different locations, same function.
Systematic rape and looting. The women not killed were subjected to violence; the property was taken.
Burning and erasure. The buildings were destroyed, the evidence concealed.
The fidelity suggested not spontaneous violence but deliberate replication. The operators had learned the pattern in June and were reproducing it in November. The algorithm had been debugged and was now running reliably.
A survivor named Amira reached Chad ten days after Ardamata.
She had hidden during the massacre—in a pit latrine, of all places, a hiding spot so foul that no one searched it. She had remained there for six hours, listening to the sounds above: gunfire, screaming, commands in Arabic, the crackle of fire.
When she emerged, the neighborhood was destroyed. Bodies in the streets. Buildings burning. Survivors wandering in shock.
She fled that night, joining a group that made its way to the border through routes that other survivors had traveled before. The road was quieter in November than it had been in June—most of the population had already fled—but the dangers remained.
In Adré, she found Ibrahim. He was already known in the camp as the teacher who documented testimonies. She told him what she had seen.
“It was the same,” she said. “The same as June. The same methods, the same sequence. As if they had a script and were following it.”
Ibrahim noted this. The observation confirmed what he had suspected: the genocide was not improvised but programmed. The patterns were reproducible because they had been designed to be reproducible.
The international response to Ardamata was, if possible, even more muted than the response to El Geneina.
By November 2023, the world’s attention had moved elsewhere. Ukraine was consuming most of the crisis bandwidth. The Israel-Gaza war, erupting in October, absorbed what remained. Sudan had become a forgotten emergency, visible only to specialists.
The UN issued a statement. Human rights organizations documented the massacre. Journalists wrote stories that appeared briefly and disappeared quickly. The words accumulated without effect.
The RSF observed this non-response and drew conclusions. The first massacre had been a test: would the world intervene? The world had not. The second massacre was confirmation: the pattern could be repeated without consequence.
The algorithm could scale.
Leila tracked Generation One in real time.
Her platform showed the spike in reports from the El Geneina region on November 4 and 5. The patterns matched what she had seen in June: communications disruption, fragmentary reports of violence, followed by a flood of survivor testimonies as escapees reached the border.
“It’s repeating,” she told the diaspora coordination team. “Same template, same sequence. They’re running the same operation.”
“Can we do anything?”
“What could we do in June? What can we do now? We can document. We can coordinate aid for survivors. We can’t stop them.”
The helplessness was corrosive. Leila had built systems for humanitarian coordination, but humanitarian coordination was downstream from genocide. The ERR networks could help survivors; they could not prevent the creation of survivors.
“They’re going to do it again,” she said. “Somewhere else. The pattern is established. Until something stops them, they’ll keep running it.”
The prediction was accurate. The question was only where and when.
The replication mechanism was visible in the personnel.
Some RSF units that had participated in El Geneina were deployed to Ardamata. The carriers of the genocide primitive moved from one location to another, bringing their knowledge with them.
Bashir was not at Ardamata—his unit was in Khartoum by November—but colleagues from his original Janjaweed cohort were there. They had learned the pattern together in 2003, refined it together in the intervening years, and now executed it together in the new genocide.
The units transmitted the pattern through direct participation. A fighter who had participated in El Geneina knew the sequence, the methods, the timing. He could perform the operation in a new location without additional training.
This was how the automaton propagated: through human carriers who moved through space, bringing the pattern with them, executing it wherever they were deployed.
The Generation One operation revealed improvements to the algorithm.
The RSF had learned from June. The encirclement at Ardamata was tighter, the escape routes more thoroughly blocked. Fewer people escaped compared to June, proportionally.
The execution was faster. Eight hours for Ardamata compared to several days for El Geneina. The efficiency had improved.
The erasure was more thorough. More bodies burned, fewer left exposed. The satellite evidence would be harder to interpret.
The algorithm was being optimized through iteration. Each generation taught lessons that improved subsequent generations. The genocide was subject to the same improvement dynamics as any other complex operation.
Ibrahim documented Generation One from the distance of Chad.
He could not witness Ardamata directly—he was already in exile—but he collected testimonies from survivors who arrived after him. Each testimony added to his understanding of the pattern.
“They’re learning,” he told Khadija. “Each time, they get better at it. More efficient. More complete.”
“Then what’s the point of documenting? If they just keep improving?”
“Because learning works both ways. We learn too. We learn the pattern, we learn where the gaps are, we learn how people escape. The next time—”
“You think there will be a next time?”
“I know there will be a next time. The question is whether we’ll be ready.”
Ibrahim’s documentation was not merely historical. It was predictive—studying the pattern to anticipate future iterations, identifying vulnerabilities that might be exploited.
The automaton was optimizing its genocide. The counter-system needed to optimize its resistance.
Generation One demonstrated that Generation Zero was not an anomaly.
El Geneina could have been explained as a particular conjunction of circumstances—local commanders acting without central direction, tribal conflicts escalating beyond control, the chaos of war producing atrocity.
Ardamata disproved these explanations. The same pattern, five months later, in the same region, with the same methods. The explanation could only be systematic: a genocide operation designed at a level above the individual massacre, deployed repeatedly according to plan.
The cellular automaton model fit. Generation Zero seeded the pattern. Generation One was the first iteration. Subsequent generations would follow—different locations, same rules.
The question was not whether more massacres would occur. The question was how many, and where, and how the counter-system would respond.
The automaton was running. The pattern was propagating. Generation Two was already being prepared.
Definition: A wave in cellular automata is a pattern that propagates through the grid while maintaining coherent structure. The Khartoum wave was different from the Darfur wave—urban terrain, mixed population, contested control. The same underlying rules produced different surface dynamics when the medium changed.
Khartoum is three cities in one.
Khartoum proper, on the southern bank of the Blue Nile. Omdurman, on the western bank of the White Nile. Bahri, north of the confluence. Together they form a metropolis of seven million people—or formed, before the war reduced that number through death and displacement.
The urban terrain shaped the propagation differently than the rural terrain of Darfur. In Darfur, the RSF could encircle villages, control access roads, achieve complete dominance over isolated populations. In Khartoum, the density prevented encirclement. Every neighborhood connected to other neighborhoods. The front lines were porous, shifting, impossible to seal.
Maryam understood this through the geography of the wounded.
In her first months at Ibrahim Malik Hospital, the casualties came from everywhere—one day from Bahri, the next from the industrial district, then from the market areas. The fighting moved like a living thing, probing for weakness, flowing around resistance.
“You can’t map it,” an older surgeon told her. “By the time you draw the line, the line has moved.”
This was the Khartoum wave: not a clean wavefront advancing across empty terrain, but a turbulent flow through complex topography.
The RSF entered Khartoum on April 15, 2023, the first day of the war.
Their initial objectives were strategic: the airport, the military headquarters, key government buildings. The assumption was that rapid decapitation would end resistance quickly—seize the centers of power, and the periphery would capitulate.
The assumption failed.
The SAF was not decapitated. Command structures relocated. Soldiers retreated to defensible positions and held. The population, instead of surrendering, organized. The Resistance Committees that had coordinated the 2019 revolution transformed into emergency response networks.
By May, the RSF controlled significant territory—perhaps 40% of the metropolitan area—but could not consolidate. The SAF counterattacked. The ERR networks sustained the civilian population in contested zones. The wave had hit a medium that resisted its propagation.
Aisha’s operation exemplified the counter-wave.
Her supply network had grown from seven hospitals in April to thirty-two facilities by June. The coverage wasn’t comprehensive—the RSF-controlled areas were inaccessible—but in contested and SAF zones, the ERR presence was nearly universal.
The network operated on simple logistics. Identify what was needed where. Identify what was available where. Match supply to demand. Transport.
“We’re running a parallel city,” she told Samira during one of their coordination meetings. “The official systems collapsed—no government services, no regular markets, no utilities. We’re replacing all of it, piece by piece.”
The replacement was imperfect. The ERR couldn’t restore electricity or running water. They couldn’t rebuild destroyed buildings or clear rubble from streets. But they could ensure that people had food, that the wounded reached hospitals, that information flowed between neighborhoods.
The counter-wave propagated through these connections. As the ERR networks expanded, their capacity expanded. Each new node added volunteers, resources, information. The network effect compounded.
Tariq, the conscripted engineering student, experienced the urban wave from the RSF side.
His unit had been assigned to hold a neighborhood in southern Khartoum—a residential area of modest houses and small shops, the kind of place where working-class families had lived for generations before the war forced them out.
The assignment was tedious and frightening in alternation. Days of nothing—sitting at checkpoints, searching the occasional vehicle, watching empty streets. Then sudden violence—a SAF probe, a mortar attack, gunfire from hidden positions.
The urban terrain favored the defenders, whoever the defenders happened to be. Buildings provided cover. Alleyways enabled ambush. The RSF had more firepower, but firepower was less decisive in streets than in open desert.
“We’re fighting ghosts,” a fellow soldier complained. “They shoot, they disappear. We can’t catch them. We can barely see them.”
The SAF’s urban tactics were effective precisely because the terrain neutralized the RSF’s advantages. The technicals that dominated Darfur were vulnerable in city streets—a single RPG from a concealed position could destroy a vehicle worth more than a year of a fighter’s salary.
Tariq began to hope that the war might end differently than he had feared. Perhaps the RSF would not win. Perhaps the urban terrain would bleed them until they negotiated. Perhaps he might survive to return to his engineering studies.
The hope was fragile, but it persisted.
The wave dynamics in Khartoum created enclaves.
RSF-controlled zones. SAF-controlled zones. Contested zones where control shifted daily or hourly. The cellular automaton’s states were not uniform; they were patchy, interleaved, a mosaic rather than a gradient.
Leila could see this patchwork in her data. Her platform tracked ERR activity by neighborhood, and the activity patterns revealed the underlying territorial control. High ERR activity meant SAF or contested—the volunteers could operate. Low ERR activity meant RSF—access was denied.
“It’s like a checkerboard,” she observed to her team. “They control some squares, we control others. No one controls everything.”
The checkerboard was unstable. Squares flipped as military operations succeeded or failed. A neighborhood that was safe on Monday might be contested on Wednesday, RSF-controlled by Friday.
The instability created humanitarian chaos. Aid delivered to a location might be inaccessible days later. Volunteers who operated freely in one zone might be captured if they strayed into another. The geography of help was as volatile as the geography of war.
The ethnic rule operated differently in Khartoum than in Darfur.
Khartoum’s population was mixed—Arab and non-Arab, Darfuri and Nile Valley, dozens of ethnicities living in proximity. The RSF could not apply the simple targeting rule that worked in Darfur’s ethnically homogeneous villages.
Instead, the targeting became more complex. Specific neighborhoods with high Darfuri populations were treated differently than neighborhoods with mixed or Arab-majority populations. Individual screening supplemented categorical violence.
Maryam noticed this in the casualty patterns. Patients from certain neighborhoods showed evidence of systematic violence—execution-style wounds, signs of torture, indicators of targeted rather than random killing. Patients from other neighborhoods showed combat injuries—shrapnel, gunshot wounds consistent with crossfire.
The distinction revealed the ethnic rule’s adaptation. Where the target population could be identified geographically, geographic targeting applied. Where the target population was dispersed, individual identification was required.
The adaptation made the urban genocide slower, less efficient, more labor-intensive than its rural counterpart. But it did not prevent the genocide from operating. It merely changed the implementation.
The SAF’s recapture of territory began in late 2024.
The momentum shifted for reasons that were partly military, partly economic, partly political. The SAF received new supplies from allies who had decided to back them. The RSF’s supply lines from Darfur were stretched. The population’s resistance, sustained by the ERR networks, had never broken.
In January 2025, the SAF launched a major offensive. Within weeks, they had pushed the RSF out of central Khartoum. By March, most of the capital was under SAF control. The RSF retreated to peripheral positions, their dream of rapid victory definitively ended.
Tariq’s unit was among those that retreated.
The withdrawal was chaotic—not a planned evacuation but a scramble to escape encirclement. Vehicles were abandoned when fuel ran out. Equipment was left behind. Some soldiers surrendered; others fled toward Darfur on foot.
Tariq found himself at a crossroads. He could continue with the RSF, retreating west into territory they still controlled. Or he could disappear—shed his uniform, blend into the civilian population, pretend he had never been a soldier.
He chose to disappear.
The uniform went into a trash pile. The rifle he had hated carrying went with it. He walked into a neighborhood he didn’t know, knocked on a door, and asked for help.
The family that answered was Arab—his ethnic category, the category that had supposedly defined his allegiance. They looked at him, saw a young man in civilian clothes, and asked no questions.
“Are you hungry?” the woman asked.
“Yes.”
She fed him. She gave him clothes that fit better. She did not ask where he had come from or what he had done.
The nafeer primitive operated even for those who had served the genocide machine. Solidarity was not conditional on deserving it.
The recapture revealed what the RSF had done.
Mass graves in neighborhoods that had been under their control. Bodies in houses, in basements, in improvised detention facilities. Evidence of torture, execution, systematic sexual violence.
Aisha’s teams were among the first to enter some of these areas. The documentation work that Maryam had begun in the hospital extended to the streets, the buildings, the burial sites.
“We have to record everything,” Aisha told her volunteers. “Photographs, locations, witness statements. Before they can deny it. Before the evidence disappears.”
The work was traumatic. Volunteers who had sustained themselves through the war by focusing on the living now had to focus on the dead. The scale of what they found exceeded anything they had imagined.
“I knew it was bad,” one volunteer said, emerging from a house where a family had been executed. “I didn’t know it was this.”
The urban wave had left its signature. The bodies told the story of how the RSF had operated in the capital: selective targeting, systematic brutality, the ethnic rule adapted for urban conditions.
The Khartoum wave demonstrated that genocide could be resisted.
In Darfur, the RSF had achieved their objectives—populations eliminated or displaced, territory cleared, the pattern executed to completion. In Khartoum, the resistance had prevented completion. The RSF had occupied but not eliminated. They had killed but not finished.
The difference was the counter-wave. The ERR networks, the SAF defense, the population’s refusal to surrender—these created friction that slowed the genocide machine. The friction accumulated until the machine stalled.
“Khartoum proves it’s possible,” Leila argued in a diaspora meeting after the recapture. “The genocide isn’t inevitable. It can be stopped. It can be reversed.”
“Only in Khartoum,” someone countered. “Darfur is still lost.”
“Then we learn from Khartoum. We study what worked. We apply it elsewhere.”
The analytical frame was correct. The Khartoum wave had been different—different terrain, different resistance, different outcome. Understanding the differences could inform strategy for future confrontations.
But the automaton was still running.
The RSF’s retreat from Khartoum did not end the war. It concentrated RSF forces in Darfur, where their control was uncontested. The gold kept flowing. The weapons kept arriving. The genocide continued in El Fasher, in Zamzam, in the camps where displaced populations had gathered.
The Khartoum perturbation had shifted the equilibrium, but not fundamentally. The system was damaged but not broken. The rules still operated. The pattern still propagated.
Maryam, surveying the wreckage of the hospital’s annex, understood this clearly. They had survived. They had won, in the limited sense that the city had not fallen completely. But the war was not over.
“What happens now?” a nurse asked her.
“We rebuild what we can. We document what we find. We prepare for what comes next.”
“You think they’ll come back?”
“I think the war will continue somewhere. Whether it comes back here, I don’t know. But somewhere, the wave is still moving.”
The cellular automaton processed. The Khartoum wave had reached a boundary, reflected, and dissipated. But other waves were propagating, other cells were transitioning, other iterations of the pattern were executing.
Part III continued. The wavefront dynamics were not finished.
Definition: The counter-wave propagates through the same medium as the primary wave but with opposite valence. Where the genocide wave destroys, the ERR wave sustains. The two waves interact, interfere, and partially cancel. The counter-wave cannot stop the primary wave, but it can reduce its amplitude, slow its propagation, and preserve nodes that would otherwise be eliminated.
The Emergency Response Rooms did not exist on April 14, 2023.
By April 20, there were twelve across Khartoum. By May, over a hundred. By June, the network had spread to every major city in Sudan where conflict had not already rendered organization impossible.
The growth was explosive, organic, self-organizing. No central committee designed the expansion. No hierarchy directed the proliferation. The pattern replicated through imitation, adaptation, and the universal recognition that coordination was necessary.
Samira watched the network grow with a mixture of pride and terror. Pride because the system worked—because the framework she and others had developed was propagating, helping people, saving lives. Terror because the need was growing faster than the capacity, because every new ERR that formed revealed ten more neighborhoods without coverage.
“We’re running to stand still,” she told Aisha during one of their increasingly rare face-to-face meetings. “Every volunteer we train, ten more are needed. Every kitchen we open, a hundred families are waiting.”
“Then we train faster. We open more.”
“There’s a limit. We don’t have infinite resources. We don’t have infinite people.”
But the limit proved farther than Samira expected. The nafeer primitive, once activated, drew on reserves that no one had measured. Volunteers appeared from populations that had seemed depleted. Resources materialized from communities that had seemed impoverished.
The wave had momentum.
The propagation mechanism was social rather than territorial.
The genocide wave spread through geographic adjacency—RSF forces moving from controlled areas to contested areas, the state transition propagating across the map. The ERR wave spread through relationship adjacency—a coordinator in one neighborhood contacting friends in another neighborhood, the organizational pattern propagating across social networks.
The two propagation mechanisms had different strengths and limitations.
Geographic propagation was limited by distance, terrain, and military capacity. The RSF could only move as fast as their vehicles, could only control as much territory as their forces could garrison.
Social propagation was limited by trust, communication, and organizational capacity. The ERRs could only spread as fast as relationships could be activated, could only coordinate as much as their communication infrastructure allowed.
In Khartoum’s dense social networks, the ERR wave propagated faster than the genocide wave. The RSF’s territorial advances were outpaced by the ERR’s organizational advances. Before the RSF controlled a neighborhood, the ERR was often already there.
Leila’s platform became a coordination backbone.
The diaspora data system, originally designed for tracking aid and displacement, evolved into a real-time operational tool. ERR coordinators from across Sudan logged into the platform, updating their status, reporting needs, coordinating responses.
“We’ve become a central nervous system,” Leila observed in July 2023. “Information comes in from everywhere. Decisions go out. The whole network operates through this hub.”
The centralization was paradoxical—the ERRs were supposed to be decentralized, autonomous, resilient against single points of failure. But some functions required coordination that only a central system could provide. Matching supply to demand across hundreds of neighborhoods required visibility that no individual node possessed.
Leila tried to maintain the balance: central coordination for information, local autonomy for action. The platform told coordinators where resources were available and where needs were greatest. The coordinators decided how to respond in their own contexts.
“Think of it like a market,” she explained to new volunteers. “The platform is the price signal. It tells you what’s scarce and what’s abundant. But you decide what to buy and sell.”
The market metaphor was imperfect—the ERRs operated outside monetary exchange, on the basis of solidarity rather than profit. But the information dynamics were similar. Distributed systems required information sharing to achieve efficient allocation.
The wave had nodes, connections, and carriers.
The nodes were the ERRs themselves: physical locations, usually private homes or community spaces, where coordination happened and resources were stored. Each node served a neighborhood, roughly defined—perhaps five thousand people in dense urban areas, fewer in less populated zones.
The connections were the communication channels: WhatsApp groups primarily, but also phone trees, runner networks for areas where phones didn’t work, any mechanism that could transmit information between nodes.
The carriers were the volunteers: the people who staffed the nodes, made the decisions, did the physical work of aid delivery and patient transport and supply logistics. Each carrier embodied the counter-primitives: identify need, coordinate response, deliver assistance, adapt to disruption, document everything.
The wave’s anatomy was visible to anyone who looked. It was not a secret organization, not a conspiracy, not a hidden structure. It operated in plain sight, advertised its presence, invited participation.
This openness was strategic. Visibility attracted volunteers, attracted donations, attracted international attention that provided some protection. The ERRs were harder to suppress when everyone knew what they were doing and why.
The interference between waves created complex dynamics.
Where the genocide wave was strong—RSF-controlled areas—the ERR wave was suppressed. Volunteers were arrested, killed, or forced to flee. Resources were looted. The counter-pattern could not propagate through territory the primary pattern controlled.
Where the ERR wave was strong—SAF or contested areas with robust networks—the genocide wave was partially dampened. The population’s resilience delayed collapse. The information flow enabled evacuation before massacres. The medical infrastructure kept casualties alive who would otherwise have died.
The interaction produced zones of different character. In some areas, the genocide was nearly complete—populations eliminated, the ERR presence destroyed. In other areas, a kind of equilibrium emerged—ongoing violence, but also ongoing resistance, neither wave fully dominant.
“We’re fighting a war of positions,” Samira said, borrowing military language. “We can’t defeat them in the field. But we can hold territory—not military territory, but social territory. We can maintain the networks where we’re strong.”
The positional warfare was exhausting. It required constant effort to maintain what existed, to rebuild what was damaged, to extend into areas that became accessible. But it was sustainable in ways that direct confrontation was not.
The ERR wave propagated beyond Sudan’s borders.
The diaspora communities became nodes in the extended network. Cairo, where hundreds of thousands of Sudanese refugees gathered. Nairobi, Addis Ababa, Kampala. London, Minneapolis, Dubai.
Each diaspora community had its own ERR structure—smaller, oriented toward remote support rather than direct service, but connected to the same network. Funds raised in Minneapolis flowed through Leila’s platform to coordinators in Omdurman. Medical advice from doctors in London reached nurses in besieged El Fasher.
“We’re one network,” Leila insisted at a diaspora coordination meeting. “It doesn’t matter where we are physically. What matters is that we’re connected, that we share information, that we coordinate action.”
The global extent gave the network resilience. The RSF could not reach the diaspora nodes. No matter what happened inside Sudan, the external nodes would survive, would continue operating, would maintain the pattern for future replication.
This was the strategic value of distributed systems: they lacked a center that could be destroyed. Kill any node, and the others continued. The network degraded gracefully rather than catastrophically.
The wave also propagated through time.
The ERR networks documented their methods, codified their procedures, trained new volunteers in systematic ways. The knowledge accumulated and transmitted, creating institutional memory that outlasted individual participants.
“We’re building something that will outlast the war,” Aisha told a group of trainees in late 2024. “The war will end eventually—one way or another. But the networks we’re building, the skills we’re teaching, these will persist. They’ll be useful for the next crisis, and the next.”
The temporal propagation was insurance against the worst outcomes. Even if the genocide succeeded, even if the RSF controlled all of Sudan, the pattern of resistance would survive in exile. The knowledge would persist. When conditions changed, the network could reconstitute.
This was the survivor strategy: maintain the seed population, preserve the genetic information, wait for conditions that would allow regrowth. The ERR wave could not win the immediate conflict, but it could ensure that the capacity for future conflict was preserved.
The interaction between waves produced selection effects.
The genocide wave eliminated populations that could not resist—isolated, unsupported, lacking the ERR infrastructure that provided early warning and evacuation assistance. The populations that survived were those with strong network connections.
Over time, this selection strengthened the overall network. The weak nodes were destroyed; the strong nodes persisted. The surviving population was increasingly organized, increasingly connected, increasingly capable of collective action.
“We’re being forged,” Ibrahim observed in his documentation. “The fire is burning away what’s fragile. What remains is harder, stronger.”
The observation was accurate but uncomfortable. The strengthening came through suffering. Each destroyed node represented real people, real deaths, real loss. The evolutionary metaphor obscured the human cost of selection.
But the pattern was real. The ERR wave, eighteen months into the war, was more robust than it had been at six months. The weak points had been found and eliminated by the enemy. The remaining structure was battle-tested.
The wave had limits.
It could not stop the genocide. It could not prevent massacres or save populations that were militarily overwhelmed. It could provide humanitarian assistance and documentation, but these were responses to atrocity, not prevention of atrocity.
Samira confronted this limit repeatedly. Families she had helped would be killed when their neighborhood fell. Supply chains she had built would be destroyed when routes were cut. The work felt like bailing a sinking ship—necessary, but not sufficient.
“We’re not winning,” she told a journalist who interviewed her in late 2024. “We’re not even holding ground, really. We’re just making the losing slower. Making sure that something survives.”
“Is that worth it?”
“What’s the alternative? Give up? Let them destroy everything without resistance? At least this way, people survive who would otherwise die. At least this way, there’s a record. At least this way, we’re not complicit through inaction.”
The moral argument for resistance was not dependent on victory. It was dependent on values—on the conviction that helping was right regardless of outcome, that documentation mattered even if justice never came, that solidarity was valuable in itself.
The ERR wave and the genocide wave continued propagating in parallel.
Neither could eliminate the other. The genocide wave had military superiority; the ERR wave had social depth. The genocide could destroy what the ERR built, but the ERR could rebuild faster than the genocide could destroy.
The equilibrium was dynamic, shifting, unstable. Territory changed hands. Networks grew and contracted. The balance of forces fluctuated with military operations, with international attention, with the internal politics of both sides.
But the structural fact remained: two waves in the same medium, propagating according to different rules, neither capable of complete dominance.
“This is what the war looks like,” Leila wrote in a report to donors. “Not a single front line, but two overlapping systems. We’re inside them, trying to strengthen ours, trying to weaken theirs. The outcome isn’t determined. It’s still being written.”
The cellular automaton processed. Both waves propagated. The interference continued.
Generation Two was being prepared—Zamzam, April 2025—and the pattern would iterate again.
Definition: Generation Two is the second iteration of a pattern propagating through a cellular automaton. It emerges from the conditions left by Generation Zero and Generation One, inherits the rules that were established, and applies them to new substrate. Zamzam was Generation Two: the same algorithm, the same executors, a different population of victims.
Zamzam was the largest camp for internally displaced persons in North Darfur.
It had been established in 2004, after the first Darfur genocide, to shelter survivors of that catastrophe. By 2025, it held approximately 400,000 people—a city of the dispossessed, technically temporary, functionally permanent.
The camp had its own geography: sectors identified by letters and numbers, pathways worn into the sand by twenty years of foot traffic, markets that had grown organically around resource distribution points. The inhabitants had built lives within the constraints of displacement—schools, clinics, mosques, the institutional infrastructure that human communities generate wherever they settle.
They had survived one genocide. They believed, perhaps, that survival conferred some protection. They were already displaced; what more could be taken?
They learned what more could be taken on April 11, 2025.
The RSF approached from the west, from positions they had consolidated after the fall of El Geneina and Ardamata.
Maryam, in El Fasher’s South Hospital, received the first reports from contacts in the camp. “They’re coming,” a coordinator messaged. “Heavy vehicles on the road. Many technicals.”
The siege of El Fasher was in its fifteenth month by then. The city was surrounded, but not yet fallen. The camp was outside the siege perimeter, theoretically separate, theoretically protected by its civilian status.
The RSF did not recognize civilian status.
The assault began at dawn. The western sectors received fire first—mortar rounds, heavy machine guns, the overwhelming violence that the RSF used to freeze populations in place.
The inhabitants could not flee. The camp was surrounded before most knew it was under attack. The roads out were blocked. The desert offered no shelter.
The pattern of El Geneina and Ardamata repeated.
Separation: men pulled from families, gathered in open areas. The ethnic rule applied—Masalit, Fur, Zaghawa identified and sorted, others sometimes released.
Execution: the men taken to locations outside the camp, wadis where the killing could occur systematically. The shooting continued throughout the day.
Looting: the shelters ransacked, the markets stripped, the twenty years of accumulated possessions taken or destroyed.
Burning: the shelters set alight, the smoke rising over the camp in a column visible for miles.
The algorithm was unchanged. The parameters were different—more people, more area, more time—but the procedure was identical.
The UN Office for the Coordination of Humanitarian Affairs documented the outcome:
Over 1,000 civilians killed in a three-day period. The true number was likely higher—many bodies were burned or buried in unmarked graves.
An estimated 400,000 people displaced, fleeing the camp that had been their refuge, scattered into the desert or toward El Fasher or toward the distant Chad border.
“One of the largest forced displacement events of the conflict,” the report concluded. The language was measured, bureaucratic, inadequate.
Leila read the report in London and felt the familiar mixture of rage and helplessness. The data on her platform showed the Zamzam node going dark—the ERR contacts there ceasing communication, the supply chain severed, the coordination structure destroyed.
“They’re doing it again,” she told the diaspora meeting. “Same pattern. They’re just working through the population systematically.”
“Can we do anything?”
“We can document. We can support survivors. We can’t stop them.”
The limitations of the counter-wave were stark. The ERR network had no capacity to prevent military assault. They could respond to the aftermath, but the aftermath was already catastrophe.
Survivors reached El Fasher over subsequent days.
They arrived on foot, in small groups, traumatized and dehydrated. The South Hospital received the wounded; the ERR networks distributed food and water; the population of the besieged city increased by tens of thousands.
Maryam treated patients whose injuries told the story. Gunshot wounds from the initial assault. Burns from the fires. The specific injuries of sexual violence that she had learned to recognize and document.
“Where is there to go?” one survivor asked her. “We fled the villages in 2003. We were in the camp for twenty years. Now the camp is destroyed. Where do we flee to?”
Maryam had no answer. The cellular automaton was processing. Each generation destroyed the refuge created by the previous generation. The survivors of Generation Zero had gathered in camps. Generation Two destroyed the camps. Where would the survivors of Generation Two gather?
The question pointed toward a terminal state: a population dispersed beyond the capacity for community, atomized into individual refugees with no collective structure. This was the goal, perhaps—not merely killing, but the destruction of the social fabric that made coherent response possible.
Ibrahim, in Chad, documented Zamzam from testimonies of survivors who reached Adré.
The testimonies echoed what he had recorded from El Geneina and Ardamata. The same methods, the same sequence, the same outcomes. The interviewees described experiences that matched patterns in his notebooks—the separation procedure, the ethnic identification questions, the destinations of the trucks.
“It’s a system,” he told a researcher from Amnesty International who visited the camp to collect evidence. “They’re running the same operation over and over. The locations change. The dates change. The victims change. The operation is constant.”
“Can you prove that? Can you demonstrate systematic coordination?”
“I can show you testimonies from three different massacres that describe identical procedures. I can show you RSF units that participated in multiple operations. I can show you the pattern.”
The proof was in the pattern. Random violence would not produce such consistency. Only organized violence, directed by common doctrine and shared training, could replicate with such fidelity.
The international response to Zamzam was familiar.
Statements of condemnation. Calls for investigation. Resolutions in appropriate bodies. No intervention, no enforcement, no material consequence for the perpetrators.
“The UN is useless,” Samira declared flatly in a coordination call. “They document. They issue reports. Nothing changes.”
“What would change things?”
“I don’t know. Military intervention? Crippling sanctions on the gold trade? Prosecution of the commanders? None of that is happening. None of it will happen.”
The fixed point of international inaction was stable. Perturbations—new massacres, new evidence, new demands—did not shift the equilibrium. The system returned to its default state: verbal condemnation without material action.
The perpetrators had learned this fixed point. Zamzam was conducted with the confidence of impunity. The RSF knew that the UN would condemn and do nothing. They knew that the gold trade would continue. They knew that no external force would intervene.
The knowledge enabled Generation Two as it had enabled Generations Zero and One.
But the ERR networks persisted.
The Zamzam coordinators had been killed or scattered, but the pattern had propagated. Survivors who reached El Fasher found ERR infrastructure waiting. Survivors who reached Chad found ERR networks already operational.
The destruction of one node did not destroy the network. The redundancy that Leila had built, the distribution that had been strategy and necessity, absorbed the loss.
Within weeks of the massacre, new ERR nodes formed among Zamzam survivors. The coordinators who had escaped trained others in the methods they had learned. The pattern replicated, adapting to new conditions, finding new substrate.
“We can’t be destroyed,” Samira said, though she meant it as aspiration rather than certainty. “They can kill us individually. They can destroy our locations. But the network regenerates. As long as there are Sudanese who remember how to help each other, the network exists.”
The nafeer primitive was resilient. The cultural substrate was harder to burn than the physical shelters.
The Generation Two propagation pointed toward Generation Three.
The pattern was established: RSF consolidation, massacre, displacement, repetition. The next target was predictable—El Fasher itself, the last major population center in Darfur not yet fully controlled.
Maryam understood this as she treated Zamzam survivors. The camp had been attacked because it was adjacent to El Fasher, because destroying it weakened the siege’s opposing forces, because the population was a resource the enemy wanted to deny.
When the siege ended, when El Fasher fell, the same pattern would execute on a larger scale. The 1.5 million people in the city would experience what 400,000 in Zamzam had experienced. The algorithm would process them through the same rules.
“They’re coming for us next,” she told Dr. Fatima—who would be killed by a drone strike three weeks later. “Everything they did at Zamzam, they’ll do here. We’re just waiting our turn.”
“Then we use the time. We prepare. We document. When it happens, we’re ready.”
“Ready for what? We can’t fight them.”
“Ready to survive. Ready to help others survive. Ready to be witnesses.”
The preparation was limited—what could civilians prepare against military assault?—but it was not nothing. The patterns Ibrahim had documented, the survival strategies that had worked elsewhere, the knowledge accumulated through eighteen months of war—these were resources that might make the difference between total annihilation and partial survival.
Generation Two demonstrated the refinement of the algorithm.
Zamzam was larger than Ardamata, faster than El Geneina. The RSF had optimized through iteration. The logistics were smoother, the coordination tighter, the outcomes more efficient.
The counter-system also refined. The ERR response to Zamzam was faster than the response to previous massacres. The documentation was more complete. The survivor support networks activated more quickly.
Both systems were learning. Both were improving. The competition was not static but dynamic—each side adapting to the other, developing countermeasures, seeking advantage.
“It’s an arms race,” Leila observed. “They get better at killing. We get better at surviving. Neither side stands still.”
The arms race metaphor was grim but accurate. The cellular automaton was not merely running; it was evolving. The rules themselves were subject to modification as the actors learned from experience.
Generation Three was coming. October 2025. El Fasher.
The pattern would iterate again, at larger scale, with whatever refinements both sides had developed. The outcome was not predetermined, but the trajectory was clear.
The automaton continued processing.
Definition: Wavefront collision occurs when two propagating patterns meet in the same medium. The interaction may produce interference, cancellation, reinforcement, or complex dynamics that neither pattern would exhibit in isolation. Where genocide and resistance collide, the outcomes are not predictable from either pattern alone.
The collision happened everywhere, continuously.
Every neighborhood in contested Khartoum was a collision point. Every village where RSF forces met ERR networks was a collision point. Every hospital where doctors treated casualties from both sides was a collision point.
The wavefronts were not neat lines on a map. They were interpenetrated, overlapping, tangled. The genocide operated in one room while the resistance operated in the next. The executioner and the helper might pass each other in the street.
Maryam lived at the collision point. Her hospital received patients regardless of their role in the war. SAF soldiers, RSF fighters, civilians caught between—all were treated according to medical need, the triage protocol that did not recognize political categories.
This neutrality was itself a form of collision. The medical primitive—heal the wounded—collided with the execution primitive—kill the targeted. When an RSF fighter was saved by a surgeon who had also treated his victims, the two primitives occupied the same space without resolution.
The collision produced boundary phenomena.
At the edges where RSF control met SAF control, the rules of both systems partially applied. Civilians in these zones navigated dual authorities, obeying whoever was locally dominant, switching allegiance as the front line shifted.
Some people exploited the boundaries. Smugglers moved supplies through the gaps between RSF checkpoints. Information couriers carried messages through contested zones. The ERR networks identified the permeable points and routed their operations through them.
Other people were destroyed by the boundaries. A family crossing from one zone to another might be killed by either side—by the RSF for fleeing, by the SAF for suspected collaboration. The boundary was both opportunity and lethal danger.
Tariq had crossed such a boundary when he deserted, shedding his RSF identity and emerging in SAF territory as a civilian. The crossing had worked—he had survived—but he knew others who had tried and failed. The boundary was selective, allowing some through and destroying others.
The collision at El Fasher was sustained for 920 days.
The siege was itself a collision phenomenon—two wavefronts pressing against each other, neither able to advance, the pressure building along the interface. The RSF pushed inward; the garrison and population pushed back. The equilibrium held until October 2025.
During those 920 days, the collision was visible in every aspect of the city’s life. The ERR networks distributed supplies under bombardment. The medical staff treated casualties while shells fell on the hospital. The population maintained social structures—schools, markets, religious observance—even as the siege tightened.
“We’re living in the collision,” Yusuf observed to Ismail, during one of the quieter periods. “Neither wave has won. Neither can advance. We exist in the space between them.”
“Can that space hold?”
“Until something changes. Until one side breaks.”
The collision was dynamic equilibrium—not stasis but sustained interaction, the two forces continuously pressing against each other. When the equilibrium collapsed, the wave with greater pressure would advance.
The collision created emergent phenomena that neither side intended.
The RSF’s siege produced a humanitarian crisis, but the crisis generated ERR infrastructure that would not have existed otherwise. The pressure created the response that resisted the pressure.
The ERR networks’ survival capacity prolonged the siege, but prolonging the siege allowed the RSF to optimize their approach. The defense enabled the refinement of the offense.
“We’re locked in a system,” Leila observed, analyzing the dynamics from London. “Every action by one side produces a reaction from the other. The reactions feed back. The whole thing becomes self-sustaining.”
The systems dynamics were those of any complex conflict: adaptation, counter-adaptation, escalation, equilibrium. But the stakes made the dynamics lethal. Each cycle of adaptation produced more refined killing, more refined survival, and more people caught between.
Individual lives were the medium of collision.
Bashir had killed, and the people he killed had been helped by the ERR networks, and the ERR volunteers had been traumatized by what they witnessed, and the trauma shaped how they helped others. The chains of causation connected every actor.
Maryam had treated RSF wounded, and the RSF wounded had returned to fighting, and their fighting had wounded others who Maryam then treated. Her medical practice was implicated in the cycle, unavoidably, regardless of her intentions.
Ibrahim documented the genocide, and his documentation might someday contribute to prosecution, and the possibility of prosecution might shape RSF behavior—or might not. The causal chains extended into uncertain futures.
No one stood outside the collision. Everyone was medium.
The collision had mathematical properties.
The interference between waves could be constructive or destructive. Constructive interference amplified effects—violence meeting resistance produced escalation. Destructive interference cancelled effects—when the genocide wave met strong ERR presence, its impact was reduced.
The pattern of interference depended on phase relationships. Where the two waves were in sync—both operating at full intensity in the same location—the collision was violent, producing the worst outcomes. Where they were out of phase—one weak where the other was strong—the collision was more survivable.
“We’re trying to maintain phase separation,” Samira explained to volunteers. “Where they’re strong, we go quiet. Where they’re weak, we push. Don’t collide directly—survive to help another day.”
The strategy was guerrilla logic applied to humanitarian work. Avoid the enemy’s strength. Concentrate on his weakness. Live to fight, or help, again.
The collision at Zamzam had been brief and decisive.
The RSF wave arrived with overwhelming force. The ERR presence in the camp was destroyed within hours. The interference lasted only long enough for the genocide wave to pass through, leaving destruction in its wake.
Brief collisions favored the stronger wave. The RSF could concentrate force, strike quickly, withdraw before effective resistance organized. The ERR networks’ strength was in duration, in persistence, in the accumulation of effects over time.
“They can’t hit everywhere at once,” Aisha observed. “So they hit one place hard, then move. Our job is to exist everywhere they’re not, and to return when they leave.”
The strategy was spatial—be where the enemy isn’t. But space was finite, and the enemy’s mobility meant nowhere was safe for long.
The collision shaped both waves.
The RSF learned to target ERR infrastructure. Coordinators were specifically identified and killed. Supply caches were located and destroyed. The genocide wave evolved to suppress the counter-wave.
The ERR learned to hide, to distribute, to maintain redundancy. Multiple coordinators for each area. Distributed supply storage. Communication security that made identification harder.
“We’re adapting to each other,” Leila wrote in her analysis. “They hunt us. We evade. They develop new hunting methods. We develop new evasion methods. The collision is accelerating both evolution curves.”
The co-evolution was visible in the changing tactics on both sides. The genocide of 2025 was more sophisticated than the genocide of 2023. The resistance of 2025 was more sophisticated than the resistance of 2023. Both had learned from the collision.
The collision would reach a climax at El Fasher.
The final assault on October 26, 2025 would be the most intense collision of the war. The strongest RSF force against the last major resistance. The genocide wave at full power against whatever remained of the counter-wave.
Maryam knew this was coming. She prepared in the ways available to her: hiding places mapped, evacuation routes identified, documentation secured in multiple locations. The collision would be overwhelming, but perhaps not total. Perhaps some would survive.
“When they come,” she told her remaining staff, “scatter. Don’t try to hold the hospital. Don’t try to resist. Hide, wait, survive. When it’s over, we’ll reassemble.”
The instruction was acknowledgment that the collision would favor the genocide wave. The counter-wave could not stop the assault. It could only preserve fragments through the cataclysm and attempt to reconstitute afterward.
This was the strategy of seeds: survive the winter, sprout in spring.
The collision was not resolution.
Neither wave could eliminate the other. The genocide could destroy nodes but not the network. The resistance could preserve survivors but not stop the killing. The collision was ongoing, perpetual, without clear endpoint.
“There’s no victory condition,” Yusuf said, in one of his last conversations before the fall. “Not for either side. We can’t win. They can’t finish. The collision continues until something external changes.”
The external change might be international intervention, economic collapse, political transformation, exhaustion of resources. Any of these might alter the collision dynamics. None were predictable. None were controllable by the actors inside the collision.
The automaton processed. The waves propagated. The collision continued.
Part III was ending. The wavefront dynamics had been examined. The propagation was understood.
Part IV would examine how the patterns reproduced themselves—how genocide begat genocide, how resistance begat resistance, how the algorithm replicated across time and space.
The collision would continue in new forms.
Definition: Self-replication is the capacity of a pattern to produce copies of itself. The copies may be identical or varied, but they preserve the essential structure that defines the pattern. The 2003 Darfur genocide was the template—the original pattern that would replicate through subsequent generations. Understanding the template is essential to understanding the copies.
Khadija was twenty-seven years old in 2003.
She lived in a village called Terbeba, in West Darfur, with her first husband and their two young children. The village was Masalit, like most settlements in that region—a community of farmers and traders who had occupied the land for generations.
The war was abstract then—news from distant places, radio broadcasts about insurgency and government response. The rebels, the government forces, the tribal militias mobilizing in response to both. None of it seemed directly relevant to Terbeba, which was small, remote, uninvolved in politics.
She learned otherwise on March 17, 2003.
The horsemen arrived at dawn. Perhaps a hundred of them, mounted, armed with rifles and swords. They came from the east, from the direction of government-controlled territory. They were not wearing uniforms, but their coordination suggested organization beyond tribal raid.
Khadija’s husband, Ahmed, told her to take the children and hide. There was a wadi behind the village, a seasonal streambed where the brush was thick. “Stay there until I come for you,” he said.
She never saw him again.
The pattern that would become standard was already present.
The horsemen surrounded the village, cutting off escape routes. They gathered the population in the central area—the market space where people traded grain and livestock. They separated men from women.
Khadija, hiding in the wadi with her children, could hear but not see. The sounds were terrible: screaming, gunfire, the distinctive thud of machetes striking flesh. Her children—a boy of four and a girl of two—did not understand. She held them, covered their ears, whispered prayers.
The sounds continued for hours. When they finally stopped, smoke rose from the village. The horsemen were burning what remained.
She stayed hidden until nightfall. When she emerged, she found Terbeba destroyed. The houses burned. The bodies lying where they had fallen—mostly men, some women, a few children. Ahmed among them.
She did not search for survivors. She took her children and walked west, toward Chad.
The 2003 genocide killed an estimated 300,000 people.
The numbers were disputed—some estimates were higher, some lower—but the scale was massive. Hundreds of villages destroyed across Darfur. Millions displaced. An entire region transformed from settled communities into a landscape of ruins and refugee camps.
The perpetrators were the Janjaweed—Arab militias recruited and armed by the government in Khartoum, deployed against the Fur, Masalit, and Zaghawa populations that were suspected of supporting rebel movements.
The methods were those Khadija had witnessed: encirclement, separation, execution, looting, burning. The methods were not improvised; they were standardized. The same pattern appeared across hundreds of villages, executed by different units but following the same template.
This was the original pattern. The pattern that Bashir learned. The pattern that the RSF inherited. The pattern that would replicate in 2023 and beyond.
Khadija survived. Many did not.
She reached Chad after two weeks of walking. She carried her daughter most of the way; her son walked beside her, too young to fully understand but old enough to know that something terrible had happened.
The refugee camp was chaos—thousands of people arriving daily, the humanitarian organizations overwhelmed, no shelter, insufficient water, disease spreading through the crowded conditions. But it was safety of a kind. The Janjaweed did not cross the border, at least not then.
She stayed in the camp for three years. Her children grew. She met Ibrahim, who had fled from a different village but the same catastrophe. They married—a practical arrangement initially, solidarity between survivors, that deepened into genuine partnership.
When the camps began to empty—when some displaced people attempted to return, when international attention drifted elsewhere—Khadija and Ibrahim moved to El Geneina. Not home, but close enough to maintain some connection to what had been lost.
For twenty years, she lived with the memory. The sound of machetes. The smoke rising. Ahmed’s body, which she had not been able to bury properly. The pattern was inscribed in her nervous system, triggered by sounds and smells that resembled that morning.
She recognized it immediately in April 2023.
The news of the war’s beginning was not surprising—Sudan had been politically unstable for years, the tensions between the army and RSF well-known. But the specific reports from West Darfur were not merely political.
“They’re killing Masalit,” she told Ibrahim. “The same way.”
“It’s been twenty years. Maybe this is different.”
“I know what I’m hearing. The same villages. The same methods. They’re doing it again.”
Ibrahim wanted to believe she was wrong—that the trauma of 2003 was distorting her perception, that the new conflict was different in kind. But the evidence accumulated. El Geneina. Ardamata. The same patterns, the same targets, the same outcomes.
The template was replicating.
The replication was not accidental.
The Janjaweed of 2003 became the RSF of 2013. The same fighters, the same commanders, the same institutional memory. The transformation was organizational—militias absorbed into formal paramilitary structure—but the methods were preserved.
Hemedti had been a Janjaweed commander in 2003. His rise to lead the RSF was not despite his genocide experience but because of it. The skills that made him effective then made him valuable later. The government that had deployed the Janjaweed continued to value their capacity.
The pattern was transmitted through human carriers. Bashir had learned from older fighters who had participated in 2003. He had transmitted what he learned to younger fighters who joined later. The knowledge propagated across generations of the organization.
“They trained us the same way,” Bashir would later tell an interviewer, years after the war ended. “The methods were passed down. Everyone knew what to do because everyone had been taught by someone who had done it before.”
The counter-pattern also had roots in 2003.
The ERR networks were novel, but the nafeer primitive they drew on was ancient. The same cultural substrate that enabled collective response in 2023 had enabled collective response in 2003—refugees helping each other in the camps, communities sharing resources during displacement, survival achieved through solidarity.
Khadija had experienced this directly. The women in the Chad camp who helped her care for her children. The men who shared food when her supplies ran low. The collective infrastructure that emerged despite the chaos.
“We survived because we helped each other,” she told Ibrahim. “That’s what we did then. That’s what they’re doing now.”
The two patterns—genocide and resistance—had both originated in the same historical moment. They had co-evolved, each shaping the other, each developing new forms in response to the other’s evolution.
The fidelity of replication was remarkable.
Researchers who studied both genocides noted the consistency: same regions, same targets, same methods. The time interval—twenty years—had not degraded the pattern. If anything, the pattern had been refined.
“High-fidelity replication,” a genocide scholar wrote in a 2024 paper. “The organizational memory of the RSF preserved the operational template with minimal drift. This suggests institutionalization—the pattern is not dependent on individual carriers but is encoded in organizational culture.”
The scholarly language captured something important. The pattern had become self-sustaining. It no longer required the original architects; it could reproduce through any carriers who were initiated into the organizational culture.
This was the mark of successful self-replication: independence from specific originators. The pattern had achieved a kind of immortality.
Khadija, hiding in El Geneina in June 2023, was living inside a replication.
The sounds were the same. The smoke rising was the same. The procedure being executed in the streets around her was the same procedure that had killed Ahmed and destroyed Terbeba twenty years before.
She was, in a sense, experiencing the same event twice. The details differed—different village, different house, older self—but the pattern was continuous. The second instantiation of the same algorithm.
“I survived once,” she told Ibrahim as they prepared to flee. “I don’t know if I can survive again.”
“We will. We have to.”
The determination was not confidence; it was refusal to accept the alternative. The pattern wanted to kill them. They would resist the pattern.
They fled on June 9, reaching Chad on June 14. Khadija had made the same journey twenty years before. The border crossing felt like a loop closing—the same trajectory, the same destination, the same refugee status.
The pattern had replicated. So had her survival.
The 2003 template explained the 2023 genocide.
Not in the sense of justification or causation, but in the sense of mechanism. The RSF did not invent their methods; they inherited them. The genocide was not spontaneous; it was reproduction.
Understanding this changed the analysis. The question was not “why is this happening?” but “why did this pattern survive, and how might it be interrupted?”
The pattern’s survival was explained by institutional continuity. The Janjaweed became the RSF. The commanders were promoted rather than prosecuted. The methods were refined rather than abandoned. The pattern persisted because the carriers persisted.
Interruption would require breaking the continuity. Prosecuting commanders. Dismantling the organization. Disrupting the transmission mechanisms that allowed the pattern to replicate.
None of this was happening. The international community had not prosecuted the 2003 perpetrators. The government had not dismantled the Janjaweed; it had institutionalized them. The pattern had been allowed to survive, and having survived, it replicated.
“This was predictable,” Ibrahim wrote in his documentation. “Everyone knew the Janjaweed had committed genocide. Everyone knew they had become the RSF. When the RSF committed genocide again, no one should have been surprised. The pattern was intact. Of course it replicated.”
The analysis was accurate but painful. The genocide was not only a crime; it was an expected outcome of allowing the template to survive. The second genocide was, in a meaningful sense, caused by the failure to prevent replication after the first.
The automaton continued processing. The pattern continued replicating. The template remained intact.
Definition: Institutional transformation is the process by which an informal pattern becomes encoded in formal structure. The Janjaweed’s transformation into the RSF was not a change of substance but a change of form—the same pattern, newly housed in official apparatus, legitimized by state recognition.
The Janjaweed had a problem: they were too effective.
By 2008, the genocide in Darfur had achieved its immediate objectives. The rebel movements were weakened. The target populations were displaced. The territory was controlled. But the methods had attracted international attention that threatened Sudan’s government.
The International Criminal Court issued arrest warrants. The United Nations imposed sanctions. The word “genocide” was officially applied. The Janjaweed had become a liability—useful for their violence, embarrassing for their visibility.
The solution was rebranding.
In 2013, Omar al-Bashir signed the decree creating the Rapid Support Forces. The RSF would be a paramilitary formation, officially integrated into Sudan’s security apparatus, commanded by Hemedti and reporting to the president. The fighters would be the same fighters. The methods would be the same methods. But the name would be different, and the formal status would provide cover.
This was the transformation: militia into paramilitary, informal into official, deniable into acknowledged.
The transformation preserved what mattered.
The RSF inherited the Janjaweed’s personnel. Bashir and thousands like him transferred directly from one organizational form to another. Their skills, their training, their experience—all preserved in the new structure.
The RSF inherited the Janjaweed’s methods. The operational template for village destruction, ethnic targeting, systematic violence—all retained and refined. The formal structure did not change the fundamental procedures; it provided better logistics for executing them.
The RSF inherited the Janjaweed’s ideology. The ethnic supremacism, the belief that Arab groups were entitled to Darfur’s resources, the willingness to eliminate obstacles—all transmitted through the personnel who carried these beliefs.
What changed was the context. The Janjaweed had operated as auxiliaries, used by the government but not formally part of it. The RSF operated as an institution, with budget lines, recruitment processes, career paths. The informality became formality.
The gold changed everything.
The Janjaweed had been dependent on government funding—unreliable, contingent on political favor, subject to withdrawal. The RSF developed independent revenue through control of gold mining operations in Darfur.
By 2017, the RSF was financially autonomous. They could pay their fighters regardless of government budget decisions. They could purchase weapons through their own supply chains. They could operate as an independent economic entity, a corporation with a military attached.
This autonomy had strategic implications. An organization dependent on government funding could be controlled by threatening to cut funding. An organization with independent revenue could not be controlled the same way. The RSF became a power unto itself, capable of operating against the government that had created it.
The transformation from Janjaweed to RSF was also a transformation from dependent to autonomous, from tool to actor.
Bashir experienced the transformation from inside.
In 2003, he had been a tribal fighter, loosely organized, paid irregularly, operating without formal structure. By 2017, he was a career soldier in a paramilitary organization, with rank, salary, benefits, and prospects for advancement.
The transformation felt like professionalization. The chaos of the early Janjaweed operations gave way to more systematic procedures. The ad hoc violence became organized violence. The tribal raiding became institutional policy.
“We became an army,” Bashir reflected later. “Before, we were just fighters. After, we were soldiers. There were rules, chains of command, training programs. It felt more legitimate.”
The legitimation was psychological as much as legal. Operating within a formal structure changed how the fighters understood their actions. They were not murderers; they were soldiers following orders. They were not war criminals; they were national security forces.
The transformation provided this psychological cover without changing the fundamental nature of the operations.
The international community’s response enabled the transformation.
The ICC warrants were never enforced. Bashir remained president until 2019, traveling internationally despite the warrants, protected by states that did not recognize the court’s jurisdiction.
The sanctions were porous. The gold trade continued despite theoretical restrictions. The UAE, in particular, provided a market for Sudanese gold that no international pressure could close.
The diplomatic engagement normalized the regime. The same government that had deployed the Janjaweed was courted for counterterrorism cooperation, for migration management, for regional stability. The crimes were tacitly forgiven in exchange for cooperation.
“The message was clear,” Ibrahim noted in his analysis. “You can commit genocide and get away with it. You can transform the perpetrators into a legitimate force. The international community will complain but will not act.”
The transformation was possible because the international system permitted it. The Janjaweed could become the RSF because no external power prevented the transformation.
The 2019 revolution tested the transformation.
When the protests began against Bashir, the RSF faced a choice: defend the regime that had created them, or align with the popular movement that demanded change.
Hemedti chose pragmatically. The RSF turned against Bashir, joining the military in deposing the president. The genocide commanders became, briefly, symbols of revolution—the forces that had toppled the dictator.
The move was strategic. Hemedti calculated that Bashir was finished and positioned himself for the post-Bashir order. The RSF’s survival mattered more than loyalty to any individual.
The revolution’s hopes were betrayed in October 2021, when the military and RSF staged a coup against the transitional government. But the brief revolutionary moment revealed something about the RSF’s character: they were not ideologically committed to any political arrangement. They were committed to their own power.
This flexibility was a feature of the institutional transformation. The RSF was not the Janjaweed, bound by tribal loyalty and government patronage. It was an independent power center, capable of allying with whoever served its interests.
The coup set the conditions for the war.
After October 2021, the RSF and SAF shared power uneasily. Both controlled armed forces. Both had economic interests to protect. The tension between them was structural—two military organizations in the same political space, neither able to subordinate the other.
The negotiations for formal integration failed. The SAF wanted to absorb the RSF; the RSF wanted to maintain autonomy. The disagreement escalated until, on April 15, 2023, it became war.
The war was not about ideology. It was about power—which military organization would control Sudan. The genocide that accompanied the war was a byproduct of which forces were fighting it. The RSF brought their methods because those methods were their methods, encoded in their institutional DNA.
“They fight the way they’ve always fought,” Yusuf observed during the siege. “The transition from Janjaweed to RSF didn’t change their tactics. It just gave them better equipment.”
The transformation demonstrated the persistence of patterns.
The Janjaweed could have been disbanded after 2003. The fighters could have been prosecuted, the organization dissolved, the methods criminalized. Instead, the pattern was preserved through transformation.
The preservation was deliberate. The government valued the capacity the Janjaweed represented. That capacity—for ethnic violence, for population control, for deniable brutality—was a resource the government did not want to lose.
The RSF was the institutionalization of that resource. The transformation was successful precisely because it preserved what the government valued while providing plausible deniability for past crimes.
“We’re not the Janjaweed,” RSF spokesmen would say when confronted with allegations. “We’re a legitimate security force.” The semantic distinction allowed the substantive continuity.
The transformation had implications for accountability.
International law struggled with the institutional form. The Janjaweed were clearly responsible for the 2003 genocide. But the RSF was a new organization, legally distinct. Could the RSF be held responsible for crimes committed by the Janjaweed, even though the personnel and methods were continuous?
The ICC’s warrants were for individuals, not institutions. Hemedti could be prosecuted for his personal role, but the organizational continuity was harder to capture in legal terms.
“The transformation is a legal firewall,” a human rights lawyer observed. “It breaks the chain of liability. The Janjaweed committed genocide; the RSF is a different entity. The criminals are protected by their own rebranding.”
The legal problem reflected a deeper conceptual problem: how to hold patterns accountable. The pattern persisted across organizational forms. The pattern was the criminal, but patterns could not be prosecuted.
The Janjaweed → RSF transformation was a case study in pattern preservation.
The lesson was clear: patterns that are not disrupted will persist. They will find new forms, new hosts, new legitimations. They will transform themselves to survive.
Disruption required more than name changes. It required dismantling the institutional structure, prosecuting the carriers, delegitimizing the methods. None of this happened after 2003. The pattern survived because it was allowed to survive.
The 2023 genocide was the consequence. The template replicated because the template was intact. The pattern self-reproduced because the reproduction mechanisms were never interrupted.
“If we want this to end,” Ibrahim wrote, “we have to break the cycle. Not just stop this genocide. Break the pattern so it can’t happen again. That means accountability. That means dismantling the organization. That means making sure the carriers can’t transmit.”
The analysis was correct but aspirational. The pattern was still operating. The carriers were still transmitting. The organization was still intact.
The transformation had succeeded. The Janjaweed had become the RSF. The pattern had found a new form.
Would it find another form after the war ended?
The answer depended on whether the disruption that hadn’t happened in 2003 would happen this time.
Definition: A constructor is an agent that builds complex systems from simple components. Hemedti was a constructor—he assembled the RSF from the remnants of the Janjaweed, the revenue from gold mining, and the political opportunities that presented themselves. His construction was a universal construction: capable of building other constructions, of scaling, of self-perpetuation.
Mohamed Hamdan Dagalo, called Hemedti, was born in 1974 in Darfur.
His origins were modest—the son of a livestock trader, from a minor subtribe of the Arab communities that occupied the northern reaches of the region. Nothing in his early life predicted the empire he would build.
But his early life included the skills that would matter. He knew the terrain of Darfur—the wadis and jebels, the water sources and trade routes. He knew the tribal politics—who aligned with whom, who could be recruited, who should be eliminated. He knew how to command men, how to inspire loyalty, how to project authority.
These skills became relevant when the government began recruiting militias to fight the Darfur rebels. Hemedti rose quickly. By 2003, he commanded significant forces. By 2007, he was one of the most powerful militia leaders in the region. By 2013, he was commander of the RSF.
The rise was not accidental. Hemedti was a constructor, and he constructed his own power systematically.
The first construction was military.
Hemedti built a personal army from the tribal networks he knew. He recruited from Arab communities, offering payment and purpose. He trained fighters in the methods that worked—the methods of the Janjaweed, refined through practice.
The army was loyal to him, not to Sudan’s government. This was deliberate. Government forces were dispersed across multiple commands, subject to political manipulation, their loyalty uncertain. Hemedti’s forces answered to Hemedti.
By 2010, he commanded perhaps 30,000 fighters. The number would grow—by 2023, the RSF had an estimated 100,000 fighters—but the core was established early. The military construction was the foundation for everything else.
“He built an army like a businessman builds a company,” an analyst observed. “He identified a market—the government’s need for deniable forces. He provided the product. He scaled up to meet demand.”
The business metaphor was apt. Hemedti approached military construction with entrepreneurial logic: identify opportunity, invest resources, grow systematically.
The second construction was economic.
The gold mining operations in Jebel Amer transformed the RSF from a dependent militia into an autonomous enterprise. Hemedti moved early to control the mines, establishing RSF dominance over the most productive areas.
The revenue was enormous—perhaps $1 billion annually by 2020. The money funded expansion, purchased weapons, paid salaries. It also funded diversification: Hemedti invested in real estate, transport, trading companies. The economic construction became a conglomerate.
“He’s not just a warlord,” a sanctions investigator explained. “He’s a CEO. The RSF is the military arm of a business empire. The gold is the core asset, but the portfolio is diversified.”
The economic construction gave Hemedti leverage that purely military commanders lacked. He could withstand funding cuts from Khartoum because he didn’t need Khartoum’s money. He could operate internationally, hiring RSF forces to Saudi Arabia and the UAE, generating additional revenue streams.
The economic autonomy was the basis for political independence. When Hemedti turned against Bashir in 2019, it was not desperation—it was calculation. He could afford to change allegiances because his power base was not dependent on government favor.
The third construction was political.
Hemedti understood that military and economic power required political cover. He cultivated relationships across Sudan’s political spectrum—with the Islamists who had supported Bashir, with the reformers who opposed him, with tribal leaders and business interests.
The relationships were instrumental. Hemedti did not have ideological commitments; he had interests. He supported whoever served those interests, abandoned whoever did not.
After the 2019 revolution, he positioned himself as a partner of the democratic transition. He appeared in public with civilian leaders, spoke the language of reform, presented himself as a guarantor of stability.
The positioning was strategic. The RSF had committed atrocities; democratic legitimacy would provide protection. Hemedti calculated that the new government would need military support, would have to negotiate with whoever controlled armed forces.
The calculation proved correct. The transitional government could not govern without military acquiescence. Hemedti made himself indispensable—and then, when the moment was right, he helped end the transition through the October 2021 coup.
The fourth construction was international.
The RSF cultivated relationships beyond Sudan’s borders. The UAE was the most important partner—a market for gold, a source of weapons, a diplomatic ally. But there were others: Saudi Arabia, Egypt, Russia.
The mercenary deployment to Yemen was part of this construction. RSF fighters participated in the Saudi-led campaign, earning revenue and building relationships. The deployment demonstrated that the RSF could project power beyond Sudan, could be useful to international partners.
“He’s building a franchise,” Leila observed, analyzing the RSF’s international network. “The same model that worked in Sudan—military force funded by resource extraction—he’s exporting it. He’s making himself useful to regimes that need deniable violence.”
The international construction provided insurance. If the RSF ever lost in Sudan, the international relationships would provide options—exile, protection, the possibility of return. The network was a hedge against total defeat.
The construction was a universal construction—capable of producing other constructions.
The RSF trained fighters who could command their own units, who could replicate the organizational model in new contexts. The financial systems created infrastructure that could support multiple enterprises. The political relationships opened possibilities that hadn’t existed before.
The construction was also self-modifying. As conditions changed, Hemedti adapted. When Bashir became a liability, Hemedti discarded him. When the democratic transition became inconvenient, Hemedti helped end it. The construction could reconfigure itself in response to circumstances.
This was the mark of a universal constructor in von Neumann’s sense: not a machine that builds one thing, but a machine that can build anything, including other universal constructors.
The genocide was one output of this construction. The same organizational capacity that built an economic empire could execute a systematic campaign of ethnic violence. The same command structure that coordinated gold mining could coordinate mass killing.
The construction did not require genocide, but it enabled it. When the war began, when the decision was made to target certain populations, the construction was capable of executing that decision at scale.
Bashir had been a carrier for Hemedti’s construction.
In 2003, he had been a fighter, learning the methods that would become the RSF’s methods. By 2023, he was a transmitter, teaching those methods to new recruits. His individual trajectory illustrated the construction’s propagation.
“I grew up in the RSF,” he said later, though the statement elided the Janjaweed origins. “Everything I know about war, I learned there. The tactics, the discipline, the way we operate—it all comes from the organization.”
The organization was Hemedti’s construction, and Bashir was a component in it. He had no independent existence as a fighter; his capacities were the capacities the construction had developed in him.
This was true for tens of thousands of RSF fighters. They were products of the construction—trained, equipped, indoctrinated by the system Hemedti had built. Their violence was the construction’s violence, executed through them but designed by the system.
The construction faced opposition.
The SAF, despite its own problems, was a rival power center that could not be absorbed. The civilian political forces, despite their weakness, represented legitimacy that the RSF could not claim. The international community, despite its inconsistency, imposed some constraints.
The war that began in April 2023 was partly a failure of the construction—the inability to achieve total dominance without conflict. Hemedti had tried to subsume the SAF through negotiation; the negotiation failed. The result was war.
But the war was also a stress test that revealed the construction’s capabilities. The RSF mobilized rapidly, seized territory effectively, sustained operations for months and then years. The construction functioned as designed, even under pressure.
“He built something durable,” Yusuf acknowledged during the siege. “I hate what they’re doing, but I respect the engineering. The RSF is a formidable organization. They don’t collapse when they take losses. They adapt and continue.”
The grudging respect reflected professional recognition. The construction was effective—not morally defensible, but operationally capable.
What happened to the constructor?
By late 2025, Hemedti’s fate was unclear. Reports placed him in Darfur, overseeing the El Fasher operation. Some claimed he was wounded; others denied it. The fog of war obscured the individual even as his construction continued operating.
The construction did not depend on his direct supervision. The systems he had built—the military command, the economic networks, the political alliances—could function without constant attention. The construction had achieved a kind of autonomy.
This was the measure of success for a universal constructor: the construction outlives the constructor. The patterns propagate independently. The system self-perpetuates.
If Hemedti died tomorrow, the RSF would not collapse. The commanders he had trained would continue commanding. The gold would continue flowing. The genocide would continue executing.
The construction had succeeded in the sense that mattered most: it had become self-sustaining.
The construction would end only through intentional dismantling.
Military defeat might destroy the RSF’s capacity, but military defeat was not certain. The SAF had recaptured Khartoum but could not project power into Darfur. The war could continue indefinitely.
Economic isolation might collapse the RSF’s funding, but economic isolation required international action that was not forthcoming. The gold trade continued. The UAE partnership continued.
Prosecution might hold the commanders accountable, but prosecution required custody, and custody required capture, and capture required military victory that might never come.
The construction had defenses against all the usual dismantling mechanisms. It was designed to survive. It had survived the first genocide, the ICC warrants, the fall of Bashir, the democratic transition. Why would it not survive this war?
“He built something that might outlast Sudan,” Ibrahim said, contemplating the trajectory. “The RSF might be the future—not a government, but a pattern that controls territory, extracts resources, persists across whatever political changes occur.”
The prospect was dark but not implausible. The construction was strong. The dismantling forces were weak.
The universal constructor continued constructing.
Definition: A self-sustaining machine is a system that generates the resources required for its own operation. The gold machine was such a system: extraction funded weapons, weapons enabled extraction, the loop closed on itself. The machine did not require external input to continue running. It was thermodynamically autonomous.
The mountain yielded approximately 80 tons of gold per year.
At global prices, this represented $4-5 billion in raw value. Not all of it flowed to the RSF—the miners took their share, the traders their cut, the government its claimed percentage. But the RSF’s portion was sufficient: perhaps $1.5 billion annually, enough to fund a private army indefinitely.
The extraction was industrial in scale if not in method. Tens of thousands of miners dug by hand, crushed rock with primitive equipment, washed the ore in basins. The technology was ancient, but the volume was modern.
Osman Idris was one of these miners. His labor at Jebel Amer was a node in the machine—human effort converted to mineral, mineral converted to wealth, wealth converted to power.
He did not understand the larger system. He knew only that he dug, and was paid, and the payment allowed him to survive. The machine’s purpose was invisible from his position within it.
The gold moved through established channels.
From Jebel Amer, the gold traveled by road to Khartoum, or directly to border crossings that led to Egypt and Libya. Some was refined in Sudan; most was exported raw or semi-processed.
The primary destination was Dubai. The UAE had positioned itself as a global gold trading hub, and Sudanese gold fit easily into the existing infrastructure. The gold arrived, was refined, was certified, was sold into international markets.
The certification process laundered the origin. Gold from a conflict zone, extracted under conditions of exploitation and violence, emerged as legitimate commodity. The transformation was alchemical in the financial sense: base metal becoming respectable asset.
“Once it’s in Dubai, it’s clean,” a gold trader explained. “The paperwork shows UAE origin. No one asks where it really came from. The market doesn’t want to know.”
The willful ignorance was profitable. The traders made commissions. The refiners added value. The banks processed transactions. The complicity was distributed across dozens of institutions, none individually responsible.
The cash returned through different channels.
Wire transfers were traceable, vulnerable to sanctions. The RSF preferred informal mechanisms—hawala networks that moved value without moving currency, cash carried by couriers, barter arrangements that avoided the formal banking system.
The money flowed to weapons suppliers. The RSF’s arsenal—the technicals, the artillery, the drones—was purchased with gold revenue. The sources were varied: old Soviet stocks, Chinese manufacturers, and increasingly sophisticated equipment that suggested state sponsors.
The weapons enabled territorial control. Territorial control included the mining areas. The mining areas produced gold. The gold purchased weapons.
The loop was closed.
Leila mapped the gold machine in late 2024.
Her analysis combined satellite imagery of mining operations, shipping data showing gold flows to UAE, weapons manifests from captured RSF materiel, and financial intelligence about RSF-connected companies.
“It’s a complete cycle,” she presented to the diaspora coordination team. “Self-sustaining. They could be completely cut off from the Sudanese government and continue operating indefinitely. The gold pays for everything.”
“Can it be disrupted?”
“In theory. Sanction the UAE refiners. Interdict the transport routes. Prosecute the traders. In practice, none of that is happening. The UAE is a strategic ally. The transport routes cross multiple jurisdictions. The traders have lawyers.”
The political economy of disruption was unfavorable. The gold machine generated wealth for too many parties. The complicity was too distributed, the interests too entrenched.
The machine had self-repair capabilities.
When one route was threatened, alternatives emerged. When one buyer faced scrutiny, others stepped in. The system was resilient because it was distributed—no single point of failure that could be targeted.
In 2023, some international attention focused on the UAE gold trade. Reports documented the RSF connection. NGOs called for action.
The response was minimal. The UAE made statements about due diligence. Some paperwork changed. The gold continued flowing.
“They adjust to pressure,” Leila observed. “They don’t stop. They just change the labels, change the routes, change the names on the documents. The machine keeps running.”
The adaptability was a feature of the machine’s design. It was built by people who expected pressure, who anticipated sanctions, who planned for disruption. The machine was hardened against the most likely attacks.
The machine generated casualties as a byproduct.
The miners at Jebel Amer worked in brutal conditions. Mercury poisoning, cave collapses, violence between competing groups. The health and safety regulations that might exist in formal mining operations did not apply.
Osman knew miners who had died—crushed by falling rock, sickened by the chemicals used in extraction, shot in disputes over promising sites. The deaths were routine, unremarked, not counted in any official statistics.
“We’re fuel,” he said once, to another miner. “They burn us to run their machine.”
The metaphor was apt. The human cost of extraction was part of the machine’s operation. The lives consumed were inputs, like diesel or explosives, necessary for the production process.
The genocide and the mining were connected at this level. Both treated human beings as resources to be consumed. Both converted lives into outputs—gold, in one case; cleared territory, in the other. The machine’s logic was consistent.
The counter-economy operated without gold.
The ERR networks were funded by diaspora contributions, not mineral extraction. The funding was smaller, less reliable, subject to the vagaries of individual giving. But it was also morally clean—no exploitation, no violence, no complicity.
“We can’t match their resources,” Samira acknowledged. “They have billions. We have thousands. But our money doesn’t kill people to get it.”
The moral asymmetry had practical implications. The ERR networks could operate transparently, could appeal to international donors, could accept funding without ethical compromise. The RSF’s funding sources, however lucrative, were criminal—subject to potential prosecution, vulnerable to reputational pressure.
The transparency vs. concealment dynamic shaped both operations. The ERRs sought visibility; the RSF sought obscurity. The ERRs publicized their work; the RSF hid their financing. The two systems operated in different registers of legitimacy.
The machine’s existence explained the war’s persistence.
Wars usually end when one side exhausts its resources. The gold machine prevented exhaustion. The RSF could sustain operations indefinitely—or at least until the gold ran out.
The deposits at Jebel Amer were substantial but finite. Geological estimates varied, but the consensus was that current extraction rates could continue for decades. Long enough that resource exhaustion was not a meaningful constraint on the war.
“We’re waiting for them to run out of money,” Yusuf said bitterly during the siege. “But they’re not running out. The gold keeps coming. The weapons keep arriving. They can besiege us forever.”
The strategic implication was grim. The SAF’s conventional resources were limited. The government in Port Sudan depended on external aid that was unreliable. The resource asymmetry favored the RSF—not because they were richer than Sudan, but because their funding was self-generating.
The machine would outlast the war.
Whatever the military outcome, the gold infrastructure would persist. The mines would continue operating. The smuggling routes would remain open. The UAE market would still exist.
If the RSF won, the machine would fund their government. If the SAF won, the machine might be captured, continuing under different management. If the war ended in partition, the machine would fund whichever entity controlled Darfur.
The machine was not attached to any particular political outcome. It was attached to the gold, which was attached to the ground, which did not care who governed.
“The gold is the only constant,” Ibrahim observed. “Governments change. Wars end. The gold remains. Whoever controls it, the machine continues.”
Breaking the machine required breaking the loop.
The obvious intervention points were the conversion nodes: where gold became cash, where cash became weapons. International action could theoretically target these conversions—sanctioning the UAE refiners, intercepting weapons shipments, prosecuting the financial networks.
But international action required international will, and international will was absent. The gold machine operated because the world allowed it to operate. The complicity was not accidental; it was structural. Too many interests benefited from the status quo.
“The machine will run until someone decides to stop it,” Leila concluded in her analysis. “Someone with power. Someone willing to accept the costs. So far, that someone doesn’t exist.”
The analysis was depressing but accurate. The gold machine was a fact of the conflict—a thermodynamic engine that would continue generating violence as long as it received fuel.
The fuel was gold. The gold was abundant. The machine kept running.
Definition: Template execution is the running of a stored pattern on new data. The El Fasher operation was the template developed in El Geneina and Ardamata and Zamzam, now executed on the largest remaining target in Darfur. The template was proven; only the scale was new.
The siege had prepared the template execution.
920 days of attrition had weakened the garrison, depleted supplies, eroded the population’s capacity to resist. The siege was not the operation itself; it was the preprocessing—the conditioning of the target data before the main algorithm ran.
By October 2025, the preprocessing was complete. The garrison was exhausted. The population was starving. The ERR networks were functioning but barely. The system was ready for final processing.
Yusuf understood this in the weeks before the fall. The mathematics had become inescapable. Supply levels were terminal. Morale was cracking. The question was not whether the city would fall but when, and what would follow.
“The siege was never the objective,” he told Ismail. “The siege was the softening. What comes after the siege—that’s what they’ve been building toward.”
What came after was the template.
The assault began on October 26, 2025.
The RSF had concentrated forces from across their controlled territory—units that had participated in El Geneina, Ardamata, Zamzam. The experienced carriers of the genocide primitive were gathered for the largest execution yet.
The approach was from multiple directions: west from their Darfur strongholds, north from positions that had encircled the city, south where the weakest defense points had been identified. The coordination was precise, the timing synchronized.
By 06:00, the northern perimeter was breached. By 08:00, RSF forces were in the city center. By noon, organized SAF resistance had collapsed. The garrison that had held for 920 days disintegrated in six hours.
The collapse was not cowardice. It was exhaustion meeting overwhelming force. The defenders had nothing left. The ammunition was gone. The will remained, but will without means was impotent.
The template executed as designed.
Phase one: encirclement and penetration. The escape routes were sealed before most civilians realized the assault had begun. The RSF controlled the roads out, the airport, the approaches to the border.
Phase two: separation. The gathering points filled with captured population—the stadium, the detention centers, the open areas that could hold thousands. Men separated from women, military separated from civilian.
Phase three: execution. The stadium massacre, where Yusuf was nearly killed, was one of many execution sites. The wadis outside the city filled with bodies. The killing continued for days.
Phase four: looting. Every structure in the captured areas was stripped. Twenty years of accumulated possessions—the things people had fled to El Fasher to protect—were taken or destroyed.
Phase five: erasure. The fires, the burning of bodies, the destruction of documents, the killing of witnesses who might later testify.
The phases overlapped, interpenetrated, ran concurrently across different sectors of the city. But the template was recognizable. Anyone who had studied El Geneina or Ardamata would have recognized the procedure.
Maryam survived by hiding.
The hospital became a death trap when RSF forces entered. The staff who remained at their posts were killed. The patients were killed in their beds. The medical neutrality that was supposed to protect healthcare facilities meant nothing.
She found a hiding place in a supply closet that had been overlooked in the initial sweep. For fourteen hours, she pressed herself against boxes of gauze and saline, listening to the sounds of murder through the walls.
The experience was dissociative. Her mind refused to fully process what she was hearing. She registered sounds—gunfire, screaming, commands in Arabic—without translating them into narrative. The narrative was too terrible; the mind protected itself through fragmentation.
When she emerged, the hospital was silent. The dead were everywhere. She walked through the destruction, documenting what she could, her phone’s camera capturing images that would later be evidence.
“I survived,” she told herself, the words a mantra. “I have to survive. Someone has to witness.”
The template’s scale was unprecedented.
El Geneina had killed perhaps 15,000. Ardamata, 1,000-1,300. Zamzam, over 1,000. El Fasher, by early estimates, killed 30,000-50,000—possibly more, as many bodies were burned or buried in mass graves that would take years to excavate.
The numbers were difficult to verify. The communications blackout prevented real-time reporting. The survivors were scattered, traumatized, unable to provide systematic accounts. The satellite imagery showed mass graves and burning, but could not count individual dead.
“We won’t know the real number for years,” Leila said in a diaspora emergency meeting. “Maybe never. They’re destroying the evidence as they kill. The erasure is part of the operation.”
The scale was not a change in the template but an application at larger scale. The same procedure that had processed thousands could process tens of thousands. The algorithm scaled.
The template also processed resistance.
The ERR networks in El Fasher were specifically targeted. Coordinators whose names were known were hunted. Supply caches that had sustained the siege population were destroyed. The counter-system was attacked as systematically as the population.
Some ERR volunteers escaped. They had prepared for this—hidden documents, alternate identities, evacuation routes. The preparation saved some lives. But many were captured and killed, their years of humanitarian work making them priorities for elimination.
“They knew who we were,” a surviving coordinator told Leila later. “They had lists. Names, photos. They were hunting us specifically. We were targets.”
The targeting revealed intelligence capacity. The RSF had identified the ERR structure, had monitored its operations, had prepared to dismantle it when the opportunity came. The genocide was not only against an ethnic population but against the organizational infrastructure that enabled survival.
Yusuf’s escape was anomaly within the template.
The template was designed to eliminate—to process every target through to destruction. His survival through the stadium chaos was a system failure, a gap in the procedure that should not have existed.
He walked for two days through desert, evading RSF patrols, navigating by stars he had learned in decades of military service. The training that had made him useful to the garrison now made him capable of escape.
He reached a village that sheltered him. He reached another village that pointed him toward Chad. The nafeer networks, operating even in territories the RSF nominally controlled, passed him from node to node until he crossed the border.
“The system leaked,” he observed later, analyzing his own survival. “They can’t seal every gap. The template has failure modes.”
The failure modes were the basis for hope, however slim. The template was powerful but not omnipotent. Some people escaped. Some witnesses survived. The automaton could not achieve perfect execution.
The template execution completed within ten days.
By November 5, 2025, the active killing operations had concluded. The city was controlled. The population was dead, fled, or detained. The template had processed El Fasher.
What remained was cleanup: continued erasure, consolidation of control, establishment of the new order. The RSF began administering the territory they had conquered, setting up the structures that would govern the depopulated city.
“El Fasher is gone,” Ibrahim said, receiving reports in Chad. “Not destroyed in the physical sense—the buildings are still there. But the city, the community, the thing that made it a place where people lived—that’s been erased.”
The observation captured something the casualty counts could not. The template did not merely kill individuals; it destroyed communities. A city of 1.5 million people—with schools, markets, hospitals, neighborhoods—was converted to empty infrastructure and mass graves.
The template would execute again.
The RSF’s territorial ambitions extended beyond Darfur. The methods proven in the west could be applied elsewhere. The template was portable, scalable, adaptable to new contexts.
Yusuf, in exile in Chad, considered the implications.
“They’ll move east,” he predicted. “They’ll move on Kordofan, on the Nuba Mountains, on anywhere they want to control. The same template, new targets. Unless someone stops them.”
No one was positioned to stop them. The SAF had recaptured Khartoum but could not project power into areas the RSF controlled. The international community continued issuing statements without action. The template execution would continue.
The El Fasher template execution demonstrated mastery.
The RSF had refined the algorithm through multiple iterations. They had scaled it to unprecedented size. They had executed it against a prepared defense that had held for 920 days. The template was robust, adaptable, effective.
The mastery was terrible to contemplate. A genocide operation so refined that it functioned as routine. A procedure so standardized that it could be deployed with industrial efficiency. A pattern so well-replicated that each new execution improved the template.
“They’ve perfected it,” Maryam said, in one of her first communications after escaping. “Whatever you call what they’ve learned to do—they’ve perfected it. They can do this to any city, any population. They’ve shown that.”
The showing was itself a message. The demonstration effect worked on multiple audiences: on populations that might resist (resistance is futile), on international observers (intervention is too late), on the perpetrators themselves (this is possible, this is permitted, this is success).
The template had executed. The pattern had replicated. The automaton continued processing.
Definition: A singular point is a moment where the system’s behavior changes discontinuously. October 26, 2025 was such a point for El Fasher—the day when 920 days of siege resolved into massacre, when potential became actual, when the pattern that had been building finally executed.
The day began before dawn.
At 04:30, RSF artillery opened fire on the northern defense positions. The barrage was heavier than any previous bombardment—not harassment fire but preparation for assault.
Yusuf was awake, as he had been awake for most of the night. The intelligence suggested something was different—unusual radio traffic, movement of forces that seemed coordinated rather than routine. He had ordered heightened alert, for whatever that was worth given the garrison’s depleted state.
The first shells landed among positions he had fortified months ago, when fortification still seemed meaningful. The positions had been designed to channel attackers into kill zones. But kill zones required ammunition to work, and the ammunition was nearly exhausted.
“This is it,” he said to Ismail. “Whatever happens today, this is the end of something.”
The northern perimeter fell faster than expected.
The RSF assault came in waves—first the infantry probes, testing defenses, then the technicals with mounted guns, then the full-scale advance when the defense points collapsed.
The garrison’s response was ragged. Units that should have provided covering fire had nothing to fire. Soldiers fell back from positions they had held for months because holding required resources they no longer had.
By 06:00, the first RSF units were inside the city. By 07:00, the fighting had become urban—street by street, building by building, the kind of combat that should have favored defenders but didn’t when the defenders were exhausted and out-supplied.
Yusuf tried to coordinate the defense from his command post, but coordination assumed communication, and the communication networks were failing. The radios that worked reported disaster: this position overrun, that unit retreating, enemy forces advancing everywhere.
Maryam’s hospital received the first casualties at 05:15.
The wounded from the initial artillery strike arrived by any means available—carried by comrades, stumbling on their own, transported in vehicles that dodged the intensifying fire. The intake was chaotic, the triage impossible.
She worked as she had worked through 920 days of siege, treating what she could, documenting what she couldn’t save. But the scale was different. The casualties exceeded anything she had seen—dozens arriving every hour, then hundreds.
By 08:00, the hospital itself was under fire. Shells landed in the courtyard, in the parking area, finally in the building itself. The surgical wing was hit at 08:47, killing three surgeons and ending any possibility of organized medical response.
“We have to go,” Dr. Fatima’s replacement said—Dr. Fatima had been killed in the March drone strike. “The building isn’t safe.”
“Where?” Maryam asked.
There was no answer. The city was falling. Nowhere was safe.
The separation operations began before noon.
RSF forces that had secured neighborhoods began the process that was now routine: gathering populations, sorting by category, routing to appropriate processing points.
The efficiency was practiced. The fighters knew the procedure. The civilians, many of whom had survived previous operations elsewhere, recognized what was happening.
Some tried to hide. The hiding places that had been prepared during the siege—basements, rooftop tanks, false walls—now became life or death decisions. Those who reached them might survive. Those who didn’t were gathered.
The gathering points filled throughout the day. The stadium, which Yusuf had fortified as a potential defensive position, became a detention site. The schools, the market squares, the open areas—all converted to holding facilities for the separated.
Yusuf’s command post fell at 10:23.
The RSF approach was methodical. They identified the building from its communications signatures, surrounded it, demanded surrender. When surrender was refused—what was the point of surrender when surrender meant death?—they assaulted.
The last defenders fought with whatever they had. Pistols when the rifles were empty. Bayonets when the pistols were empty. Some fought hand-to-hand until they were overwhelmed.
Yusuf was captured alive. The RSF recognized his rank, his value as a trophy. A colonel who had commanded the 920-day defense was not to be killed immediately—he was to be transported, processed, perhaps interrogated.
They bound his hands and put him in a vehicle. The vehicle joined a convoy heading toward the stadium.
The executions began in the afternoon.
The stadium, filled with separated men, became the processing center. The procedure was assembly-line: groups moved from the field to the far goal, where the trenches waited. The shooting was continuous, a background rhythm to the day.
Yusuf was held separately from the general population—the officers were kept in a section near the entrance, presumably for different treatment. He could see the processing, could count the groups moving toward the goal, could hear the shooting that ended each group.
He calculated: at the current rate, perhaps four hours until his group would be processed. Four hours of watching others die, knowing his turn was coming.
The calculation was interrupted by the SAF airstrike at 21:30. The bomb—part of a last-ditch effort by forces in Port Sudan—hit near the stadium, close enough to cause chaos among the guards.
Yusuf ran. He did not decide to run; his body ran. The gap in the fence, the darkness, the confusion—all of it aligned into an escape route that he took without conscious thought.
Maryam found her hiding place at 09:00.
The supply closet in the hospital’s lower level had been her refuge during previous bombardments. It was cramped, dark, accessible only through a corridor the RSF might not search thoroughly.
She took water, her phone, her documentation—the photographs and recordings she had accumulated through months of siege. If she survived, the documentation would matter. If she didn’t, perhaps the phone would be found someday, the records recovered.
The fourteen hours she spent in hiding were the longest of her life. The sounds filtered through—gunfire, explosions, screaming, commands. She pieced together what was happening from the fragments, constructed a narrative of massacre from acoustic evidence.
At 23:00, the sounds faded. The hospital was quiet. She waited until 05:00 the next morning before emerging, until the silence had lasted long enough to suggest the immediate operations were complete.
What she found confirmed what she had heard.
The city’s transformation was visible in the streets.
Bodies. Buildings burning or burned. The infrastructure of civilian life—shops, homes, vehicles—destroyed or abandoned. The silence that wasn’t silence but the absence of normal sound, the negative space where a city’s noise should be.
Maryam walked through this landscape, documenting. Her phone’s battery was low, but she photographed what she could—the bodies, the destruction, the scale of what had happened. The documentation primitive operating even in extremity.
She encountered other survivors—people emerging from hiding places, searching for family members, trying to orient themselves in the transformed city. Some joined her; others fled in different directions. The ERR networks, fragmented by the assault, began attempting to reconnect.
October 26 ended, but its effects continued.
The killing operations extended through October 27, 28, 29. The separation and execution template ran until the target population was processed. The erasure operations—burning, burial, destruction of evidence—continued longer still.
The day was a singular point, but singular points have durations. The discontinuity wasn’t instantaneous; it was a phase transition that took days to complete.
By early November, the phase transition was finished. El Fasher had changed state: from besieged city to captured territory, from living community to genocide site. The automaton had processed through its transition.
October 26, 2025 would be remembered.
It would be the date on memorials, the anniversary of commemoration, the historical marker that divided before from after. The 920-day siege would be remembered through this ending—the defense that failed, the city that fell, the population that was massacred.
But remembrance required survivors, and survivors required escape, and escape required the system’s imperfection. Yusuf escaped through chaos. Maryam survived through hiding. Others survived through luck, through help, through the gaps in the template that could never be perfectly sealed.
“We remember,” Ibrahim said, in a ceremony held in the Chad camps on the first anniversary. “We remember the dead. We remember the living. We remember what happened so that it cannot be denied.”
The remembrance was itself a form of resistance. The erasure rule wanted October 26 forgotten, the evidence destroyed, the history erased. The documentation rule preserved it—in testimonies, in photographs, in the memories of survivors.
The singular point had occurred. The automaton had processed. But the record remained.
Definition: Replication fidelity measures how accurately a copy reproduces the original pattern. High fidelity means the copy is nearly identical; low fidelity means significant variation. The Sudan genocide exhibited remarkably high replication fidelity—each iteration closely matching the template, with only minor adaptations.
The researchers measured fidelity through pattern comparison.
They aligned the testimonies from El Geneina (June 2023) with testimonies from Ardamata (November 2023) with testimonies from Zamzam (April 2025) with testimonies from El Fasher (October 2025). They coded the events for procedural elements: encirclement methods, separation protocols, execution techniques, looting patterns, erasure practices.
The correlations were striking. Across four years and four major operations, the procedural elements matched at rates exceeding 90%. The same methods, the same sequence, the same organizational signatures.
“This level of fidelity requires explicit transmission,” a genocide studies professor wrote. “You don’t get 90%+ correlation through independent invention. This is trained behavior, institutional doctrine, copied from generation to generation.”
Ibrahim’s documentation had anticipated this finding. His notebooks contained side-by-side comparisons that showed the consistency. He had recognized the replication before the academics quantified it.
The fidelity was maintained through carrier transmission.
Bashir had learned the methods in 2003, refined them through subsequent operations, transmitted them to fighters he trained. The same was true for hundreds of other experienced carriers. The pattern propagated through person-to-person teaching.
The teaching was not formal—there were no classrooms, no manuals, no written doctrine. The transmission was embedded in operations. A new fighter joined a unit, participated in operations alongside experienced fighters, absorbed the methods through practice and observation.
“You learn by doing,” Bashir explained, though his explanation came years later, after circumstances had changed. “The first operation, you watch. The second, you assist. By the third, you know what to do.”
The learning was bodily, procedural, encoded in muscle memory and automatic responses. The fighters didn’t think about the methods; they enacted them. The fidelity was high because the transmission was embodied rather than abstract.
Variation occurred at the margins.
The core procedures were stable, but details varied by operation. Local terrain influenced tactics. Commander preferences shaped timing. Resource availability affected scale.
At El Geneina, the executions were distributed across multiple sites—wadis, open areas, locations convenient for burial. At Ardamata, the executions were more concentrated, processed through fewer sites. At El Fasher, the stadium became a central processing facility that hadn’t existed in earlier operations.
The variations were adaptations to local conditions, not changes to the fundamental pattern. The template remained constant; the implementation details adjusted.
“Same algorithm, different parameters,” Leila observed, using her computational frame. “They’re running the same program. The inputs differ, so the outputs differ. But the transformation is identical.”
The counter-pattern also exhibited replication fidelity.
The ERR methods propagated with similar consistency. The coordination protocols developed in Khartoum appeared in El Fasher. The documentation practices refined after El Geneina were applied at Zamzam. The pattern of resistance copied itself as effectively as the pattern of genocide.
Samira was a primary carrier of the counter-pattern. The methods she had developed in the 2019 revolution, adapted for humanitarian response, propagated through the volunteers she trained. Her students became trainers themselves, transmitting the methods to subsequent generations of ERR workers.
“We’re both replicating,” she observed in a moment of dark reflection. “They copy their methods. We copy ours. The war is a competition between replications.”
The parallel was uncomfortable but structurally accurate. Both systems achieved high fidelity through the same mechanisms: embodied learning, person-to-person transmission, refinement through practice.
Fidelity degradation would eventually occur.
All replication systems experience error. Copies accumulate mutations. Over time, the copies diverge from the original.
In biological systems, this divergence produces evolution—new forms emerging from accumulated variations. In social systems, it produces drift—procedures changing gradually as carriers interpret them differently.
The genocide pattern had not yet experienced significant drift. The 22-year gap between 2003 and 2025 had not degraded the methods. The institutional memory of the RSF had preserved the template with remarkable accuracy.
But the preservation depended on continuous transmission. If the carriers were eliminated—prosecuted, killed, dispersed—the transmission would stop. The pattern would lose fidelity as it passed through fewer and fewer experienced hands.
“Break the transmission,” Ibrahim argued in his analysis. “That’s how you stop the replication. You have to sever the links between generations of perpetrators.”
The fidelity measurement was itself evidence.
The consistency across operations demonstrated coordination. Random violence would produce variation; systematic violence produced consistency. The high fidelity proved that the genocide was organized, directed, intentional.
“This is our strongest evidence,” a prosecutor at the ICC explained. “The pattern matching shows command responsibility. Someone is designing these operations. Someone is ensuring they execute consistently. That someone bears criminal liability.”
The evidentiary value depended on documentation. Ibrahim’s notebooks, Maryam’s photographs, Leila’s data—all contributed to establishing the pattern. The fidelity that proved genocide required witnesses who could testify to the consistency.
Khadija recognized the fidelity immediately.
She had survived 2003. When the 2023 operations began, she knew what she was seeing—not similar violence, not comparable atrocity, but the same thing. The pattern that had killed her first husband and destroyed her first community was executing again.
“It’s identical,” she told Ibrahim, before they fled. “The same sequence. The same methods. They’re doing exactly what they did before.”
Her recognition was based on embodied memory. She didn’t need statistical analysis to measure fidelity; she had lived through the template. Her body remembered the original pattern, and the copy triggered the same responses.
The bodily recognition was testimony of a particular kind. It couldn’t be quantified, couldn’t be entered into academic databases, but it was evidence nonetheless. The survivors knew what they knew.
The fidelity had limits.
Perfect replication was impossible. Each operation introduced some variation, some adaptation that became part of the pattern for subsequent operations.
The drone warfare that developed through the siege of El Fasher was not part of the 2003 template. The drone hadn’t been used then; the technology had evolved. But the RSF incorporated drones into their operations, and the drone strikes became part of the pattern.
The counter-pattern also evolved. The diaspora coordination platforms that Leila built didn’t exist in 2003. The hawala networks had always existed, but their systematic integration into humanitarian response was new. The ERR methods absorbed innovations that improved their effectiveness.
“Both patterns are evolving,” Leila noted. “They’re not static copies. They’re adaptive systems, incorporating new capabilities while maintaining core structure.”
The evolution was consistent with high fidelity—the core remained stable while the periphery adapted. The template persisted; the implementation improved.
What would break the fidelity?
The prosecution of carriers would disrupt transmission. The experienced perpetrators who transmitted the methods would be removed from the system. New carriers would have fewer teachers, weaker transmission, lower fidelity.
The death of carriers would have similar effects—whether through military action, natural attrition, or the aging out of the generation that had learned the methods in 2003.
The institutional dissolution would destroy the organizational structure that enabled transmission. If the RSF were disbanded, the carriers would scatter. The pattern would fragment.
None of these disruptions were occurring. The carriers remained free. The institution remained intact. The transmission continued.
“The fidelity will persist as long as the system persists,” Ibrahim concluded. “They’ve built a self-perpetuating pattern. It will copy itself until something breaks the mechanism.”
The replication fidelity was the pattern’s strength and vulnerability.
Strength: the methods worked. They had been proven across multiple operations. The high fidelity meant that successful methods were preserved and transmitted. The pattern was optimized.
Vulnerability: the consistency was evidence. The pattern matching proved coordination. The fidelity that made operations effective also made prosecution possible.
The vulnerability would only matter if prosecution occurred. The evidence existed. The documentation was compiled. The pattern was established. But prosecution required political will, international action, the capture of perpetrators—none of which were happening.
“The evidence condemns them,” Maryam said. “But condemnation without consequence is just words.”
The fidelity measurement was complete. The pattern was documented. The replication was proven.
What happened next depended on whether anyone would act on the evidence.
Definition: Counter-replication is the copying of a pattern designed to oppose another pattern. The ERR networks counter-replicated: they copied themselves to match the spread of genocide, creating resistance infrastructure wherever the genocide machine operated. The counter-replication did not prevent the primary replication; it created a parallel system that persisted alongside it.
The ERRs were born in Khartoum and propagated outward.
The initial nodes formed in the capital during the first weeks of the war. By June 2023, the pattern had spread to Port Sudan, to Wad Madani, to every major city where Sudanese refugees gathered. By 2024, the pattern had crossed borders—ERR structures in Chad, in Egypt, in Uganda.
The propagation followed displacement. Wherever Sudanese fled, they carried the ERR pattern with them. The coordinators who had operated in one location became nuclei for new nodes in their destinations.
“We’re seeds,” Samira said. “Each of us who escapes carries the pattern. We plant it wherever we land.”
The botanical metaphor was apt. The ERR structure was designed for dispersal. The methods were simple enough to transmit quickly, robust enough to function with minimal resources, adaptable enough to work in varied contexts.
The counter-replication matched the genocide’s reach.
Wherever the RSF operated, ERR networks formed in response. The genocide created refugees; the refugees created ERR nodes. The violence produced the displacement that enabled the counter-pattern’s spread.
This coupling was not accidental. The ERR model was designed to respond to crisis—any crisis, anywhere. The genocide simply provided the crisis that activated the response.
“They’re making us stronger,” Aisha observed, paradoxically. “Every time they displace a population, that population organizes. The more they attack, the wider we spread.”
The observation contained a dark truth. The counter-replication depended on the primary replication for its fuel. Without genocide, there would be no refugees. Without refugees, the ERR networks would not spread. The resistance was parasitic on the violence it opposed.
The replication preserved knowledge across displacement.
The coordinators who fled El Geneina brought their experience to Chad. They knew the genocide pattern—had survived it, documented it. They knew the counter-pattern—had operated it under extreme conditions.
This knowledge transferred to new contexts. The ERR nodes in Chad incorporated lessons from El Geneina. The nodes that formed after El Fasher incorporated lessons from all previous operations. Each displacement wave enriched the collective knowledge.
“We’re building a library of survival,” Ibrahim said. “Every operation teaches us something. The pattern accumulates.”
The accumulated knowledge was the counter-system’s competitive advantage. The genocide pattern was stable, optimized, resistant to change. The counter-pattern was adaptive, learning, incorporating improvements from each iteration.
Leila’s platform enabled network-wide learning.
The coordination system she had built allowed knowledge sharing across all nodes. A technique developed in one location could be documented, transmitted, and adopted elsewhere. The platform accelerated the counter-replication’s evolution.
“We’re doing distributed research,” she explained. “Every ERR group is an experiment. What works, we share. What fails, we learn from. The whole network benefits from each node’s experience.”
The distributed research model had advantages the centralized RSF couldn’t match. The RSF’s methods were transmitted through hierarchy—from command to units, from experienced fighters to new recruits. The ERR’s methods were transmitted through network—peer to peer, horizontally, without hierarchical bottlenecks.
The network structure made the counter-replication more robust. Destroying any node didn’t destroy the knowledge; the knowledge was distributed across all nodes. The RSF could kill coordinators but couldn’t kill the pattern they had transmitted.
The counter-replication had limitations.
The ERRs could spread, but they couldn’t stop the genocide. They could help survivors, but they couldn’t prevent the creation of survivors. The counter-pattern was defensive, responsive, unable to achieve victory in any conventional sense.
“We’re not winning,” Samira acknowledged in a rare moment of pessimism. “We’re just not losing completely. That’s different.”
The difference mattered. Not losing completely meant survival. Survival meant the possibility of future change. The counter-replication was a holding action, preserving capacity until conditions shifted.
But conditions might never shift. The genocide might continue until its targets were exhausted. The counter-replication might persist indefinitely in a system that produced refugees faster than it could help them.
The counter-replication created social infrastructure.
Beyond the immediate humanitarian function, the ERR networks built relationships, trust, organizational capacity. The survivors who participated developed skills in coordination, logistics, documentation. They became carriers of civic capacity as well as survival skills.
“We’re building the next generation of civil society,” a diaspora organizer argued. “These volunteers—they’re learning how to organize. When the war ends, they’ll have experience that will matter.”
The argument was optimistic but not implausible. Revolutions and reconstruction required organizational capacity. The ERR networks were developing exactly that capacity—tested under the most extreme conditions, refined through years of practice.
The counter-replication was not only spreading resistance. It was spreading the skills that might, eventually, build something better.
The fidelity of counter-replication varied by context.
The urban ERRs in Khartoum operated differently than the camp-based ERRs in Chad. The diaspora nodes had different resources and constraints than the nodes inside Sudan. The pattern adapted to local conditions more than the genocide pattern did.
This variation was both strength and weakness. Strength: the flexibility allowed operations in diverse contexts. Weakness: the coordination became harder as the variations accumulated.
Leila spent significant effort on standardization—developing protocols that could work across contexts, creating templates that preserved core functionality while allowing local adaptation.
“We need enough consistency to coordinate,” she said. “But not so much that we’re inflexible. It’s a balance.”
The balance was the central challenge of counter-replication. Too much consistency, and the system couldn’t adapt. Too little, and the system couldn’t coordinate.
The counter-replication competed with the primary replication for substrate.
Both patterns wanted the same thing: the population. The genocide wanted to eliminate the population. The ERR wanted to sustain the population. Each survivor was a victory for one side and a defeat for the other.
The competition played out in every neighborhood, every camp, every moment of the war. The genocide pattern tried to process the population through its rules. The counter-pattern tried to route the population away from processing.
“We’re fighting for the same people,” Maryam observed. “They want them dead. We want them alive. Every person who survives is someone they failed to kill and someone we succeeded in helping.”
The framing clarified the stakes. The counter-replication was not abstract resistance; it was specific humans surviving. The numbers of survivors were the measure of the counter-pattern’s success.
The counter-replication would continue until the genocide ended.
The ERR networks were not designed for permanence; they were designed for crisis. When the crisis ended—if it ended—the networks would transform into something else, or dissolve, or merge into formal institutions.
But the crisis showed no sign of ending. The genocide continued. The displacement continued. The need for counter-replication continued.
“We’ve become permanent by accident,” Aisha reflected in late 2025. “We thought this would be temporary. A few months, maybe a year. Now we’ve been doing this for over two years, and there’s no end in sight.”
The permanence was exhausting. The volunteers who had signed up for emergency response found themselves in indefinite commitment. The resources that had been mobilized for short-term crisis were stretched across extended operations.
Part IV demonstrated self-replication in both directions.
The genocide replicated: from 2003 to 2023, from El Geneina to Ardamata to Zamzam to El Fasher. The pattern copied itself with high fidelity, executing the same template across time and space.
The resistance replicated: from Khartoum to Port Sudan to Chad to the global diaspora. The counter-pattern copied itself with adaptive fidelity, spreading wherever displacement created the conditions for its propagation.
The two replications were locked in co-evolution. Each shaped the other. The genocide’s spread triggered the counter-replication’s spread. The counter-replication’s resistance influenced the genocide’s methods.
“We’re in a system,” Leila wrote in her analysis. “Two replicating patterns, interacting, evolving together. Neither can eliminate the other. Both persist.”
The system was dynamic but not resolving. The equilibrium—if it could be called that—was stable in its instability. Both patterns continued, neither dominant, neither eliminated.
Part V would examine the fixed points: the stable configurations that persisted despite the dynamic competition. The points of equilibrium in a system that was never at rest.
Definition: A fixed point is a state that remains stable under the system’s dynamics. The system may be perturbed—pushed away from the fixed point—but it returns to the same state. The international community’s response to the Sudan genocide was a fixed point: ⟲ statements → nothing → statements. No matter what happened, the system returned to this configuration.
The pattern was established before the war began.
In 2003, the world had learned of Darfur. The media had covered the genocide. Colin Powell had called it genocide. The United Nations had passed resolutions. The ICC had issued arrest warrants.
And nothing material had happened.
The pattern—condemnation without consequence, statement without action, concern without intervention—became the international community’s default response. It was not a conscious choice but an emergent property of how international institutions functioned.
“They express concern very effectively,” Ibrahim observed bitterly. “They’ve perfected the expression of concern. The concern means nothing, but they express it beautifully.”
The fixed point had structural causes.
The UN Security Council required great power consensus for meaningful action. Russia and China could veto intervention. The Western powers, even if willing, could not act alone without Security Council authorization.
The regional organizations—the African Union, the Arab League—had neither the capacity nor the will to intervene. They were composed of governments that feared the precedent of intervention in another state’s internal affairs.
The international legal institutions—the ICC, the International Court of Justice—could issue warrants and rulings, but lacked enforcement mechanisms. The perpetrators ignored them without consequence.
Each institution, individually, had reasons for inaction. Collectively, they produced a system that could not act—a fixed point maintained by the structural constraints of international order.
The 2023 war tested the fixed point.
The scale of violence was undeniable. The documentation was overwhelming. The genocide label was applied early and accurately. All the conditions that should trigger international response were present.
The response followed the familiar pattern.
The UN issued statements. The Security Council debated. The Human Rights Council convened special sessions. The language escalated—”deeply concerned,” “gravely alarmed,” “strongly condemns”—without the actions escalating correspondingly.
“The vocabulary of concern has no upper bound,” Leila noted. “They can always add another adverb. ‘Profoundly deeply gravely concerned.’ But the actions stay at zero.”
The fixed point was perturbable but stable.
Events that seemed like they should shift the system—major massacres, documented war crimes, starving children on television—did produce temporary changes. Attention increased. Pressure built. Occasionally, minor actions occurred: targeted sanctions on individuals, arms embargoes that were not enforced.
But the system returned to equilibrium. The attention faded. The pressure dissipated. The minor actions remained ineffective. The fixed point reasserted itself.
“It’s like pushing a ball uphill,” a humanitarian coordinator explained. “You can push it partway up. But if you stop pushing, it rolls back down. The stable state is at the bottom. Nothing we do changes the stable state.”
The RSF understood the fixed point.
Their strategy incorporated the assumption of international inaction. They knew statements would come but actions would not. They planned accordingly.
The timing of major operations was revealing. The El Fasher assault launched when international attention was focused on other crises—the ongoing situation in Gaza, tensions in the Taiwan Strait, the US presidential transition. The RSF attacked when the world was looking elsewhere.
“They read the news like we do,” Yusuf observed. “They know when they can move. They know the international system won’t respond. They factor that into their planning.”
The fixed point was not merely a constraint the RSF faced; it was a resource they exploited. The predictability of international inaction enabled strategic planning. The stability of the fixed point was, for the perpetrators, a form of permission.
The fixed point had been challenged before.
Libya 2011 had seemed like a departure. The Security Council authorized intervention. NATO forces supported the rebels who toppled Gaddafi. The doctrine of Responsibility to Protect—R2P—appeared to have teeth.
But Libya’s aftermath discredited intervention. The country collapsed into chaos. The intervention that had seemed like humanitarian success became a cautionary tale. The lesson learned was not “intervention can work” but “intervention causes problems.”
“Libya killed R2P,” a diplomat said. “After Libya, no one wants to be the one who proposes intervention. The precedent was negative.”
The failed precedent strengthened the fixed point. Each argument against intervention could point to Libya. The stability of inaction was reinforced by the instability that action had produced.
The victims experienced the fixed point as abandonment.
For Maryam, in the besieged hospital, the international community was an abstraction that occasionally produced supplies—when supply flights could land—and consistently produced statements that changed nothing.
“They talk about us,” she told a colleague. “We hear the statements. ‘Deeply concerned about the humanitarian situation in El Fasher.’ We’re inside the humanitarian situation. Their concern doesn’t translate to anything we can use.”
The gap between statement and experience was unbridgeable. The international system operated at a level of abstraction that had no direct connection to bodies in hospital beds, to food running out, to shells landing on buildings.
The fixed point was stable partly because the victims had no influence over it. The system responded to pressures from its powerful members, not from the populations it nominally protected.
Some actors tried to shift the fixed point.
Human rights organizations documented and publicized, hoping that enough exposure would generate political pressure. Journalists risked their lives to report from conflict zones. Diaspora activists lobbied governments, organized protests, demanded action.
These efforts produced perturbations. A particularly shocking report might generate a news cycle. A powerful photograph might briefly capture attention. A diplomatic initiative might create the appearance of movement.
But the perturbations decayed. The news cycle moved on. The attention shifted. The appearance of movement revealed itself as stasis. The fixed point remained.
“We’re Sisyphus,” a human rights researcher said. “We push the rock up. It rolls back down. We push again. It rolls back down. But we can’t stop pushing, because the alternative is accepting that nothing can change.”
The fixed point might eventually shift.
Systems that seem stable can undergo phase transitions—sudden shifts to new stable states triggered by accumulated pressure or unexpected events.
The conditions for such a shift were unclear. Perhaps if a major power decided Sudan mattered strategically. Perhaps if the genocide spread beyond Sudan’s borders, threatening regional stability. Perhaps if the RSF somehow threatened Western interests directly.
“The system will move when it wants to,” Leila said. “When something changes the cost-benefit calculation for the powerful. Not because of morality. Because of interests.”
The analysis was cynical but historically supported. International interventions had occurred when powerful states decided intervention served their interests. Humanitarian concern was rhetoric; strategic interest was motive.
The international fixed point was part of the genocide’s enabling conditions.
The RSF committed genocide partly because they could. The international system’s stable inaction was a permissive condition—not the cause of genocide, but a factor that made genocide viable.
If intervention were certain—if genocide reliably triggered military response—the RSF’s calculations would change. The fixed point of inaction enabled the fixed point of genocide.
The two fixed points were coupled. The international system’s stability enabled the genocide system’s operation. Breaking one might break both; neither could be broken easily.
“We’re stuck in an equilibrium,” Ibrahim wrote in his analysis. “The genocide continues because no one stops it. No one stops it because the system isn’t designed to stop it. The design is the problem. But no one can redesign the system while the genocide continues.”
The circular trap was the fixed point’s most pernicious feature. The system that needed to be changed was the system that prevented change. The stability was self-reinforcing.
Part V would examine the other fixed points: the economic, the humanitarian, the documentation systems that had stabilized around the conflict. Each stable configuration interlocked with the others. The compound stability was the war’s deepest structure.
Definition: The gold fixed point is an economic equilibrium that sustains itself through cyclic flows: ⟲ mine → refine → export → cash → weapons → control → mine. The loop is self-financing, self-reinforcing, impervious to external pressure as long as each link in the chain remains functional. The genocide operates because the gold fixed point provides its economic substrate.
The gold never stopped flowing.
Through twenty months of war, through massacres and sieges and international condemnation, the supply chain continued. Trucks carried ore from Jebel Amer to processing facilities. Refined gold moved to airstrips and border crossings. Chartered flights and overland convoys carried it to Dubai. Cash returned through hawala networks and wire transfers.
The monthly volume barely fluctuated. Eighty tons in April 2023. Eighty-two tons in October 2023. Seventy-nine tons in March 2024. Eighty-one tons in September 2024. The war was background noise to the extraction machinery; the killing enabled rather than disrupted the commerce.
Leila tracked these flows through fragmentary data—export records, shipping manifests, the occasional leaked financial document. The picture that emerged was of an economy operating with mechanical reliability.
“They’ve built a perpetual motion machine,” she told a investigative journalist who was preparing a story on RSF financing. “Not literally—there’s energy input from the mines, from the workers, from the fuel and equipment. But economically, it sustains itself. The gold pays for the operation that extracts the gold.”
The fixed point had crystallized over a decade.
When Hemedti’s forces first seized control of Jebel Amer in 2017, the gold trade was chaotic—artisanal miners, competing militias, fragmented supply chains. The RSF imposed order. They established monopoly control over the major mining sites. They built processing facilities. They negotiated exclusive export agreements.
By 2020, the system was mature. The extraction was industrialized. The refining was professionalized. The export channels were established and reliable. The revenue stream was predictable enough to secure loans, to fund long-term planning, to build institutional capacity.
The war in 2023 was, from one perspective, an attempt to expand this system. The RSF sought control of all Sudan, not merely Darfur. But the gold remained the foundation. Without Jebel Amer, the war could not be sustained. With Jebel Amer, the war could continue indefinitely.
Osman Idris still worked the mines.
He was fifty-four now, two years older than when the war began. His body showed the accumulation of labor—the bent spine, the calloused hands, the persistent cough from years of inhaling rock dust. He would die in these mines, he knew. The only question was when.
The war had changed nothing about his daily routine. He descended into the pit at dawn. He extracted ore with hand tools. He hauled the ore to the surface. He received payment from the traders who circulated through the camp.
What had changed was the context. The RSF’s control was tighter now. The checkpoints were more numerous. The tribute demanded from miners had increased. The profit margin narrowed while the labor remained constant.
“Before the war, I could send money to my daughter every month,” he told another miner during a break. “Now, maybe every two months. They take more. They always take more.”
The fixed point sustained itself partly by extracting from those at its base. The miners received the minimum necessary to keep them working. The surplus flowed upward—to the traders, to the RSF, to the Dubai refineries, to the weapons dealers.
Dubai was the essential node.
The UAE had positioned itself as the world’s gold trading hub—no questions asked, no sources examined, no documentation required beyond the minimal. The infrastructure existed: refineries that could process any quality of ore, banks that could move any quantity of cash, logistics networks that could reach any destination.
For the RSF, Dubai provided what no domestic economy could: conversion from physical gold to liquid capital. A ton of gold in Darfur was wealth but not power. A ton of gold converted to dollars in Dubai could purchase weapons from any market in the world.
The relationship was transactional but stable. Dubai did not care about the provenance of the gold or the purposes to which the money would be applied. Dubai cared about transaction fees, refining margins, banking charges. The genocide was irrelevant to these calculations.
“They’re not supporting the RSF,” a financial analyst explained to Leila. “They’re not opposing the RSF either. They’re simply providing services to anyone who pays. The RSF pays, so they provide services. It’s not ideology. It’s commerce.”
The distinction was morally significant but practically meaningless. The result was the same: gold flowed to Dubai, cash flowed back, weapons arrived, the genocide continued.
The weapons arrived through multiple channels.
Some came directly from state suppliers—countries that maintained formal or informal alliances with the RSF. The UAE provided drones and vehicles. Other Gulf states provided ammunition and equipment. The transactions were deniable but observable.
Some came through black market networks. Eastern European arms dealers. Chinese manufacturers selling through intermediaries. The global weapons trade was too fragmented, too poorly regulated, too profit-driven to exclude a paying customer.
Some came from captured SAF stockpiles. Each military victory yielded weapons that could be turned against the next target. The war fed itself.
The diversity of supply channels made the fixed point resilient. If one channel closed—through sanctions, through diplomatic pressure, through supplier decisions—others remained open. The weapons kept arriving because someone was always willing to sell to a buyer with gold-backed cash.
Sanctions were proposed, debated, and implemented inadequately.
The US had sanctioned individual RSF commanders. The EU had sanctioned certain RSF-linked companies. These measures created inconvenience but not interruption. The sanctioned individuals used intermediaries. The sanctioned companies reorganized under new names.
The gold trade was never seriously targeted.
To sanction Sudanese gold exports would require cooperation from the UAE, which was unwilling. It would require enforcement mechanisms that did not exist. It would require tracing chains of custody that were deliberately obscured.
“They can sanction individuals all day,” an RSF-linked businessman told a journalist. “We have a hundred individuals. Sanction one, another takes their place. But they can’t sanction gold. Gold is gold. Once it’s refined, it’s indistinguishable from any other gold in the world.”
The fungibility of gold was the fixed point’s ultimate protection. The physical commodity bore no identifying marks. Once it entered the global supply chain, it mixed with gold from every other source. The genocide’s proceeds became untraceable.
The fixed point created dependencies.
The miners depended on the RSF for access to the sites. The traders depended on the RSF for safe passage. The refineries in Dubai depended on Sudanese gold for a significant portion of their throughput. The weapons dealers depended on RSF orders for revenue.
Each participant in the chain had incentives to maintain the system. Breaking the chain meant breaking their own income. The fixed point was stable because everyone involved benefited from its continuation.
This web of dependencies extended beyond the direct participants. Governments that traded with Dubai, companies that purchased refined gold, consumers who bought jewelry containing Sudanese gold—all were implicated in the chain, however distantly.
“We’re all connected,” Leila said during a presentation to a human rights organization. “The gold in your wedding ring might have funded a massacre. The electronics in your phone might contain gold mined by displaced farmers under RSF guns. The supply chains are too long and too opaque to know for certain.”
The presentation was meant to be galvanizing. The audience response was mostly discomfort. The implications were too large, too diffuse, too entangled with ordinary life to permit easy action.
Colonel Yusuf had understood the fixed point from inside the siege.
“We’re not fighting soldiers,” he had told Ismail during one of their strategic discussions. “We’re fighting an economic system. The RSF can lose a thousand men and still continue. They can lose equipment and resupply. They can lose battles and still win the war. Because as long as the gold flows, they have unlimited capacity to regenerate.”
The SAF, by contrast, depended on government budgets—budgets that were shrinking as the war consumed resources and territory. The SAF could not regenerate. Each loss was permanent. The asymmetry was fundamental.
“To beat them, you’d have to take Jebel Amer,” Yusuf continued. “Cut the gold. Cut the money. Cut the weapons. Everything else—the battles, the sieges, the diplomatic negotiations—is secondary. The war is decided at the mines.”
The analysis was correct but useless. The SAF had no capacity to project force to Jebel Amer. The international community had no will to intervene in the mining regions. The fixed point would persist because no one with the capability to disrupt it had the interest to do so.
Some attempted to disrupt the fixed point from the edges.
Investigative journalists published exposés of the gold trade. Human rights organizations documented the connection between mining and massacre. Activists pressured retailers to audit their gold supply chains.
These efforts produced noise but not disruption. The retailers issued statements about ethical sourcing without changing their purchasing practices. The governments expressed concern without imposing enforceable sanctions. The exposés generated news cycles that faded within days.
The fixed point was too robust, too distributed, too embedded in the global economy to be broken by peripheral pressure. Breaking it would require attacking the core: either the mines themselves or the Dubai conversion node. Neither was politically feasible.
“We’re nibbling at the edges,” an activist admitted to Leila. “We know the edges don’t matter. But the core is unreachable. So we nibble, and we hope that somehow the nibbling accumulates into something significant.”
The hope was not supported by evidence. The nibbling had been ongoing for years, and the fixed point remained stable.
The gold fixed point and the international fixed point were coupled.
The international community’s inaction enabled the gold trade to continue. The gold trade enabled the genocide to continue. The genocide’s continuation demonstrated the futility of international response, which reinforced the pattern of inaction.
The coupling created a compound stability more robust than either fixed point alone. To break the cycle would require simultaneous disruption of both points—international action that actually targeted the economic base, economic pressure that actually achieved international coordination.
Neither was forthcoming. The fixed points persisted. The gold flowed. The genocide continued.
“This is what stability looks like in hell,” Ibrahim wrote in his analysis. “Not chaos. Not disorder. A stable system that produces death as its steady-state output. The horror is not that the system is breaking down. The horror is that it’s working exactly as designed.”
The observation captured the essential truth of fixed points: they were stable because they functioned. The genocide was not a failure of systems; it was a success—a horrible, efficient success—of systems optimized for extraction and elimination.
Part V continued examining these stable horrors.
Definition: The famine fixed point is a humanitarian equilibrium of sustained suffering: ⟲ siege → starve → die → siege. The system does not collapse into total death nor recover into adequate nutrition. It stabilizes at a level of chronic starvation that kills slowly, predictably, without triggering interventions sufficient to break the cycle. Famine is not chaos; it is a stable configuration of controlled deprivation.
Thirty million people were hungry.
The number was so large as to be abstract. Thirty million—roughly the population of Texas, of Peru, of Malaysia. Each of those thirty million experienced their hunger individually: the cramps, the weakness, the cognitive fog, the desperation. The abstraction concealed thirty million particular agonies.
The humanitarian system had a term for this scale: “famine conditions.” The term was clinical, calibrated to international response frameworks. Famine conditions in X districts. Famine-level food insecurity affecting Y million. The language converted mass starvation into reportable metrics.
Maryam, in the remnants of South Hospital, watched the metrics arrive in human form.
Children with kwashiorkor—their bellies distended, their hair turned orange, their skin cracking from protein deficiency. Adults reduced to skeletal frames, their bodies consuming themselves for energy. Elderly patients whose immune systems had collapsed, dying from infections that adequate nutrition would have prevented.
The fixed point was visible in their bodies: not quite dead, not adequately fed, suspended in a stable state of deterioration.
The famine was not an accident.
It was produced—deliberately, systematically, as a component of the siege strategy. The RSF blocked food convoys. They destroyed agricultural equipment. They attacked markets and distribution centers. They created the conditions under which starvation was inevitable.
This was a war crime clearly defined under international law. The deliberate use of starvation as a weapon. The prevention of humanitarian access. The targeting of civilian food systems.
The laws existed. The crimes were documented. Nothing happened.
“They’re starving us on purpose,” a camp resident told an aid worker in Zamzam, months before the camp’s destruction. “Everyone knows. The whole world knows. And still no food comes.”
The aid worker had no answer. She too knew. Her organization had reported the blockades, had documented the denial of access, had briefed diplomats and journalists. The knowledge changed nothing.
The international food aid system was also at a fixed point.
The World Food Programme operated on annual funding cycles. Donors committed amounts based on political calculations unrelated to need. The commitments were always less than required—sometimes much less.
In 2024, the Sudan appeal was 30% funded. In 2025, with El Fasher falling and the famine spreading, funding dropped further as donors diverted resources to other crises.
“There’s a finite pool of humanitarian money,” a WFP official explained to Leila. “Sudan competes with Ukraine, with Gaza, with Yemen. The donors decide what matters. Right now, Sudan doesn’t matter enough.”
The logic was zero-sum: dollars for Sudan meant dollars not available elsewhere. The humanitarian system, designed to respond to crises, had become a system for rationing response among simultaneous crises. The rationing left Sudan underfunded.
“But Sudan has the largest displacement crisis in the world,” Leila protested. “The largest number of people in famine conditions—”
“I know. It doesn’t matter. The funding follows attention, and attention follows politics. Sudan isn’t politically important to the major donors. So Sudan starves.”
The ERRs tried to fill the gap.
Their community kitchens operated on donations from the diaspora, on whatever local food could be obtained, on the labor of volunteers who were themselves often hungry. The scale was small relative to the need—hundreds of thousands reached, while millions starved.
Aisha coordinated seven kitchens in Omdurman through 2024. Each kitchen served roughly 500 people daily. Seven kitchens, 3,500 people fed. The neighborhood contained 200,000 people. The gap between capacity and need was a factor of fifty.
“We’re not solving the problem,” she told her team during one of their coordination meetings. “We’re demonstrating that a solution is possible. If we can feed 3,500, and there are a thousand groups like us, we could feed 3.5 million. But there aren’t a thousand groups like us. There aren’t enough resources for a thousand groups like us.”
The ERR model worked at neighborhood scale. It did not scale to national crisis. The nafeer primitive was powerful but not omnipotent. It could not create food from nothing.
The fixed point maintained itself through several mechanisms.
First, the blockades continued. As long as the RSF controlled the roads, food could not reach besieged populations in quantities sufficient to break the famine.
Second, the funding remained inadequate. As long as donors prioritized other crises, the humanitarian system could not operate at scale.
Third, agricultural production was destroyed. Even if blockades lifted, the farms were burned, the livestock killed, the farmers displaced. Recovery would take years.
Fourth, the displacement continued. People who might have grown food were in camps, dependent on aid that wasn’t coming. The famine displaced people; displacement worsened the famine.
Each mechanism reinforced the others. The siege created dependency on external food; external food was blocked; unblocked food couldn’t reach destroyed farms; farmers couldn’t farm because they were displaced; displacement increased the population needing external food.
The loop was closed. The fixed point was stable.
Children died at a steady rate.
The mortality data, when it could be collected, showed consistent patterns. Under-five mortality elevated by a factor of three to five above peacetime baseline. Acute malnutrition rates at 30% or higher in the worst-affected camps. Death as a statistical regularity, predictable and predicted.
Maryam stopped counting individual deaths after the first thousand. The number was meaningless. What mattered was the rate, the trend, the systematic grinding away of a generation.
“They’re not killing these children directly,” she said to a colleague. “No soldier shoots them. No bomb falls on them. They simply don’t eat, and then they die. It’s murder by omission. Murder by policy. Murder by the mathematics of food distribution.”
The colleague nodded. There was nothing to add. They had both seen too many small bodies to maintain any illusion about what was happening.
Some children survived through the ERR networks.
The kitchens prioritized children under five. The distribution systems tried to identify the most vulnerable. The volunteers carried sacks of grain to houses where mothers had stopped producing milk because they themselves were starving.
These rescues were individual, specific, inadequate. Each saved child was surrounded by children who could not be saved. The triage was brutal: this one will live, that one will not, resources permit saving only so many.
Samira made these decisions daily.
“I choose who eats,” she said during a rare moment of reflection. “Not officially—I’m not in charge of anything. But when there’s not enough, someone decides. And somehow it’s me. I point at that family, and they eat. I don’t point at that family, and they don’t.”
“How do you decide?”
“Children first. Sick children before healthy children. Pregnant women. Nursing mothers. After that...” She shrugged. “I don’t know. I just decide. And then I live with what I’ve decided.”
The weight of the decisions accumulated. The volunteers who made them carried scars that would never fully heal.
The famine was visible from space.
Yale HRL’s satellite imagery showed the signatures: abandoned agricultural land turning from cultivated patterns to scrub, IDP camps expanding as desperate populations concentrated, market areas empty where commerce had once thrived.
But satellite imagery could not show the hunger itself. The data captured consequences—migration, land use change, settlement patterns—not the experience. The abstraction remained intact even in high-resolution imagery.
“We can see where people aren’t farming,” a researcher explained. “We can see where people are congregating. We can infer food stress from these indicators. But we can’t see the hunger. The hunger is invisible to our instruments.”
The invisibility protected the comfortable. Statistics about famine-level food insecurity were distant, deniable, easy to ignore. The hungry were real, but their reality did not transmit through the channels that reached decision-makers.
The famine fixed point had a temporal dimension.
Starvation that lasted months became chronic malnutrition. Children who survived but were stunted would never recover their potential height or cognitive capacity. The damage was permanent even if the famine ended tomorrow.
“We’re not just killing people,” a nutritionist told Leila. “We’re damaging a generation. The children who survive will be shorter, weaker, less cognitively developed. They’ll have higher rates of chronic disease. They’ll die younger. The famine extends its effects decades into the future.”
This long shadow was the fixed point’s most insidious feature. Even intervention—even adequate food supply starting immediately—could not undo the damage already inflicted. The famine had written itself into bodies, into developmental trajectories, into the future Sudan would inherit.
The famine was part of the genocide.
The legal definition of genocide includes “deliberately inflicting on the group conditions of life calculated to bring about its physical destruction.” Starvation was such a condition. The famine was not separate from the massacres; it was a component of the same system.
The bullets killed quickly. The famine killed slowly. Both were expressions of the same intent: the elimination of populations marked by the ethnic rule.
Ibrahim understood this as he analyzed his documentation. The testimonies of massacre survivors and the statistics of famine deaths told the same story from different angles. The RSF had constructed a comprehensive system of elimination, employing multiple methods toward the same end.
“They shoot when shooting is efficient,” he wrote. “They starve when starving is efficient. The method varies; the purpose is constant.”
The fixed point of famine was coupled to the other fixed points: the international inaction that permitted blockades, the gold that funded the siege, the erasure that concealed the scope of the dying. The compound stability held all the fixed points in place.
Breaking one might break all. But no one was breaking any of them.
The children continued to die. The fixed point continued to hold.
Definition: The documentation fixed point is a cycle of record and futility: ⟲ witness → record → report → ignore → witness. The evidence accumulates; the response remains constant at zero. Documentation becomes an end in itself, a ritual of bearing witness that has decoupled from any mechanism of consequence.
Ibrahim had filled seventeen notebooks.
Each notebook was A5 size, 200 pages, college-ruled. He had purchased them in El Geneina before the war, intending them for lesson plans and student assessments. Instead, they had become a record of genocide.
The entries were systematic. Date. Location. Source (eyewitness, secondhand, documentary). Event description. Names when known. Details that might later serve as evidence: distinguishing features of perpetrators, specific weapons used, sequences of action.
He had developed the methodology through practice, refining his approach as he interviewed more survivors. The early notebooks were less organized; the later ones followed a consistent format that any investigator could parse.
Seventeen notebooks. Approximately 3,400 pages. Perhaps 400 discrete incidents recorded. Thousands of names—victims, perpetrators, witnesses.
The documentation was comprehensive. The documentation was useless.
The ICC had issued arrest warrants.
The International Criminal Court, following years of investigation, had charged RSF commanders with war crimes, crimes against humanity, genocide. The warrants were public. The evidence was overwhelming.
No arrests occurred.
The RSF commanders remained in Sudan, protected by their forces, beyond the reach of any enforcement mechanism. The countries that might have arrested them—the UAE, Saudi Arabia, Egypt—had no intention of doing so. The warrant was a symbol without consequence.
“We have the evidence,” an ICC prosecutor told a documentary crew. “We have the witnesses. We have the documentation. What we don’t have is the power to enforce our rulings. The Court depends on state cooperation. The states aren’t cooperating.”
The structure was familiar from previous genocides. Rwanda had eventually seen trials, but only after the genocide was complete and the regime had changed. Darfur’s 2003 genocide had produced ICC charges that remained unenforced twenty years later.
The documentation fixed point operated within this structure: careful recording, precise legal analysis, issued warrants, and then nothing.
Multiple organizations were documenting simultaneously.
The Yale Humanitarian Research Lab analyzed satellite imagery. Amnesty International conducted survivor interviews. Human Rights Watch compiled reports. The UN OHCHR gathered testimony. Physicians for Human Rights documented medical evidence. The Clooney Foundation for Justice provided legal analysis.
Each organization followed professional methodologies. Each produced reports meeting international evidentiary standards. Each published findings to media and policymakers.
The collective output was overwhelming. Hundreds of reports. Thousands of pages. Millions of words describing, analyzing, proving what everyone already knew: genocide was occurring.
“We’ve documented this more thoroughly than any previous atrocity,” a human rights researcher observed. “The Holocaust was documented in real-time, but with primitive technology. Rwanda happened too fast for comprehensive documentation. This—we have satellite imagery of every mass grave. We have thousands of survivor testimonies. We have financial records, communication intercepts, defector accounts. The evidence is complete.”
“And what does the completeness accomplish?”
The researcher had no answer.
Leila’s database contained 847,000 records.
She had built it systematically, aggregating reports from ERR coordinators, from hospital records that could be extracted, from testimonies gathered by multiple organizations. Each record represented an incident, a death, a displacement, a documented harm.
The database was searchable. It could generate statistics by region, by date, by type of incident. It could cross-reference perpetrators across multiple events. It could identify patterns that no single report would reveal.
“This is the most comprehensive data on a genocide ever assembled in real-time,” she told a conference of human rights lawyers. “We know more about what’s happening in Sudan than anyone knew about any previous mass atrocity while it was happening.”
“What do we do with the knowledge?” someone asked.
“That’s the question I can’t answer. We know. We document. We report. And then we watch it continue.”
The database grew daily. The genocide grew alongside it.
The documentation served purposes beyond legal accountability.
It preserved memory. Even if no perpetrator faced trial, the record would exist. Future generations would know what happened—not in vague terms, but in precise detail.
It countered denial. When the RSF claimed civilians weren’t targeted, the documentation contradicted them. When governments minimized the scale, the documentation corrected the record.
It supported survivors. People whose suffering was recorded felt, at least partially, acknowledged. The documentation was witness, even when it was not justice.
Ibrahim had come to understand these subsidiary purposes. His notebooks might never be admitted in court. The evidence he gathered might never lead to prosecution. But the record would exist. The names would be preserved. The dead would not disappear entirely into the RSF’s erasure.
“I document because documentation is what I can do,” he wrote in an introduction to his archive. “I cannot stop the killing. I cannot feed the starving. I cannot enforce international law. I can write down what happened. Perhaps that is enough. Perhaps it is not enough. But it is what I can do.”
The documentation was itself at risk.
Physical records could be destroyed. Ibrahim had lost one notebook during his flight to Chad—left behind in a moment of panic, probably now burned. Digital records could be hacked, erased, corrupted.
The organizations that gathered evidence maintained redundant backups, but redundancy was not invulnerability. The RSF specifically targeted documentation efforts. Journalists were killed. Human rights workers were detained. Witnesses were eliminated.
The competition between documentation and erasure was ongoing. Each new piece of evidence was contested by the effort to destroy it. The documentation fixed point held partly because documenters kept working despite the risks.
Maryam had photographed fifty-seven bodies in South Hospital on October 27, 2025. She had transmitted the images through the brief window of connectivity before the blackout resumed. The images now existed in multiple locations—on her phone, on cloud servers, in the databases of human rights organizations.
If she were killed, the images would persist. If her phone were destroyed, the copies would survive. The redundancy was deliberate, a safeguard against the erasure rule.
The futility of documentation produced despair among the documenters.
“What am I doing?” an Amnesty researcher asked during a team meeting. “I interview survivors. I write reports. The reports are published. Nothing happens. The survivors go back to camps where they might be killed tomorrow. And I write another report.”
The question had no satisfying answer. The documentation continued because the alternative—not documenting—felt worse. If no one recorded what was happening, the erasure would be complete. Documentation maintained the possibility of future justice, however remote.
“We’re planting seeds,” another researcher offered. “Maybe nothing grows in our lifetime. Maybe the trials happen in twenty years, in fifty years. But without the documentation, trials could never happen. We’re creating the conditions for future accountability.”
The metaphor was comforting but uncertain. Seeds planted in toxic soil might never grow. The conditions for future accountability might never materialize.
The documentation fixed point was coupled to the international fixed point.
Evidence was gathered to inform decision-makers. Decision-makers received the evidence. Decision-makers did not act. More evidence was gathered, on the theory that previous evidence had been insufficient.
The theory was false. The decision-makers’ inaction was not caused by inadequate evidence. Adequate evidence existed from the first month of the war. More evidence did not produce more action because action had never depended on evidence.
“They’re not waiting for proof,” Leila realized during one of her analytical sessions. “They have proof. They’re waiting for something else—political will, strategic interest, whatever threshold causes them to care. Evidence doesn’t create that threshold. Evidence just documents our approach to it.”
The insight was dispiriting. The documenters had been operating on a theory of change—document atrocity, inform decision-makers, generate response—that did not match the actual dynamics. The fixed point held because the theory was wrong.
Ibrahim continued documenting anyway.
He had found a rhythm in the camps. Each day, he conducted interviews. Each evening, he transcribed notes. Each week, he organized materials and transmitted summaries to the networks that collected them.
The routine provided structure. In a world that had collapsed into chaos, the discipline of documentation was a form of order. He controlled something, even if what he controlled was only the careful recording of things he could not change.
“You’re obsessed,” Khadija told him one night. “You spend all your time with the dead. With their stories. You’re not living anymore—you’re archiving.”
“Someone has to remember.”
“The living need you too. I need you. We survived. Don’t we deserve your attention as much as those who didn’t?”
The tension was real. Documentation had become Ibrahim’s purpose, but purpose could become escape. The fixed point of witness could trap the witness, binding them to a record of horror at the expense of present life.
He tried to balance. Some days for interviews. Some days for Khadija. Some hours for transcription. Some hours for simply being present, for existing without recording.
The balance was imperfect. The dead demanded attention. Their stories pressed against him, requiring expression. The documentation fixed point held him as firmly as it held the international community—though in his case, the stability felt more like devotion than dysfunction.
The fixed point would persist until something broke it.
Perhaps a political change would create demand for evidence. A new government, a successful prosecution, a shift in international priorities.
Perhaps the evidence would accumulate until it triggered an unexplained phase transition. Some tipping point where documentation finally produced consequence.
Perhaps the fixed point would never break. The documentation would grow into an archive, studied by historians but never used for justice.
Ibrahim could not know which future would materialize. He could only continue his contribution: one more testimony recorded, one more name preserved, one more piece added to the structure of memory.
“The documentation may never accomplish justice,” he wrote. “But without documentation, justice is certainly impossible. I do not know if my work will matter. I know only that not working is surrender. And I have not surrendered.”
The fixed point held. The witness continued witnessing. The cycle continued cycling.
The other fixed points awaited examination.
Definition: The nafeer fixed point is a positive equilibrium of mutual aid: ⟲ need → help → trust → need. Unlike the other fixed points, this one produces life rather than death. The cycle sustains itself through reciprocity: those who receive help become those who give help; trust built through giving enables future asking. Nafeer is the counter-fixed-point to genocide—stable, self-reinforcing, and opposite in its outputs.
The kitchen in Bahri had been operating for 947 days.
It had begun on April 16, 2023—the second day of the war—in Samira’s aunt’s house. Three women cooking for their immediate neighbors. Forty portions of fool, the bean stew that was Sudan’s staple, distributed to families whose breadwinners couldn’t reach their workplaces.
By the third week, the kitchen served 300 people daily. By the third month, 800. By the first anniversary of the war, over 1,500.
The scale had grown, but the principle remained constant: identify need, prepare food, distribute, adapt to changing circumstances. The nafeer primitive, implemented through the kitchen’s daily operations.
Samira had lost track of how many volunteers had participated over 947 days. Hundreds, certainly. Some had come for a few weeks, then moved on—displaced, exhausted, or needed elsewhere. Others had become permanent fixtures, their identities now inseparable from the kitchen’s identity.
“We’re not an organization,” Samira said when journalists asked about the structure. “We’re a practice. A habit. People come, people cook, people eat. The practice continues.”
The fixed point maintained itself through reciprocity.
A family received food during the first month of the war. By the third month, the father had found irregular work. They no longer needed to receive; they began to give. Small contributions—some vegetables from a relative’s garden, labor for a few hours when time permitted.
The contributions were not payment. No one owed anything specific. But the general sense of obligation—I was helped, I should help—created a flow of resources into the system that partially offset the outflow.
“It’s not charity,” Aisha explained to a researcher studying the ERR networks. “Charity creates dependents. This is nafeer. Participation is expected from everyone who can participate. We’re not helping the poor. We’re helping each other.”
The distinction was philosophically significant and practically important. Charity positioned helpers above those helped. Nafeer positioned everyone in the same network, roles shifting based on circumstance. Today I give; tomorrow I receive; the day after I give again.
The fixed point had expanded through replication.
The Bahri kitchen was one node in a network that, by late 2025, included over 900 ERR-affiliated operations across Sudan and in refugee camps in Chad and Egypt. Each node had emerged independently, following the same pattern: local organizers responding to local need, using local resources.
The nodes connected through WhatsApp groups, through diaspora coordination, through the informal communication networks that Sudanese society had developed. A technique discovered in one kitchen—a more efficient stove, a better method for distributing while minimizing crowding—spread through the network within days.
“We’re not coordinated in the traditional sense,” Samira observed. “No one gives orders. No one reports to anyone. But we’re connected. We learn from each other. We share what works.”
The network architecture resembled peer-to-peer file sharing more than hierarchical organization. Each node was autonomous; the network was emergent; the coordination was distributed.
The diaspora sustained the fixed point economically.
Leila had tracked the flows: approximately $200 million transferred through diaspora channels in the war’s first two years. The money came from Sudanese communities in the Gulf, in Europe, in North America. It arrived through hawala, through formal remittances, through cryptocurrency transfers.
Relative to the official humanitarian budget, the amount was small. The UN appealed for billions; the diaspora contributed hundreds of millions. But the efficiency was incomparable.
“The UN system has overhead,” Leila explained. “Administrative costs, logistics, security, all the infrastructure required to operate in a conflict zone. The ERRs have almost no overhead. A dollar from the diaspora becomes ninety cents of food. A dollar from the UN becomes forty cents of food, maybe less.”
The efficiency was possible because the ERRs were volunteers, using their own vehicles, their own houses, their own labor. The overhead was absorbed by the community rather than budgeted as organizational cost.
“We’re subsidizing ourselves,” Samira noted. “The volunteers aren’t paid. The facilities are donated. The fuel comes from our own pockets or from contributions. We’re converting our own resources into aid.”
The conversion was sustainable only because of the nafeer ethos. People gave their time and resources because the culture demanded it—because they had internalized the primitive that help was obligatory.
The fixed point faced constant perturbation.
RSF advances displaced the populations that ERRs served. Volunteers were killed, detained, or forced to flee. Supply chains were interrupted. The war’s chaos constantly disrupted operations.
But the fixed point was resilient precisely because it was distributed. When one kitchen closed, others absorbed its beneficiaries. When volunteers left, new volunteers emerged. The network adapted continuously, never achieving stability in any particular configuration but maintaining stability as a pattern.
“We’re not a building that can be bombed,” Samira said. “We’re a practice that can’t be bombed. Destroy one kitchen, another appears. Kill volunteers, others step forward. The practice persists because it exists in people, not in structures.”
The RSF had tried to target ERR operations. In Khartoum, they had attacked known distribution points. In Darfur, they had arrested suspected coordinators. But the attacks produced adaptation rather than elimination.
The ERR networks became more covert, distributing through shifting locations rather than fixed points. They developed communication security, using encrypted apps and code words. They built redundancy, ensuring that no single individual’s capture could compromise the network.
The fixed point was coupled to the other fixed points through opposition.
The international inaction left the ERRs as the primary humanitarian response. The gold fixed point funded the genocide that the ERRs resisted. The famine fixed point created the need that the ERRs addressed. The documentation fixed point recorded what the ERRs witnessed.
Each negative fixed point created conditions that strengthened the nafeer fixed point. As international aid failed, informal aid became more essential. As famine deepened, the kitchens became more vital. As genocide continued, the solidarity networks became more determined.
“They’re trying to break us,” an ERR coordinator said. “But their attacks make us stronger. Every family they displace becomes a family we serve. Every horror they commit becomes a reason for volunteers to join. The worse things get, the more nafeer responds.”
The dynamic was not entirely positive. The ERRs grew because the need grew. The need grew because the genocide continued. The strength of the mutual aid network was, in some sense, a measure of the scale of the catastrophe.
The ERRs had become internationally visible.
Nobel Peace Prize nominations in 2024 and 2025. The Chatham House Prize. The Right Livelihood Award. Media coverage that portrayed the volunteers as heroes in a landscape of horror.
The recognition was deserved but uncomfortable.
“We don’t want prizes,” Samira said when informed of the Nobel nomination. “We want the war to end. We want food for everyone who is hungry. We want people to stop dying. The prizes don’t help with any of that.”
The international attention did help with fundraising. The diaspora contributions increased after major media coverage. The awards ceremonies provided platforms to publicize the crisis.
But the attention also created risks. Recognition made ERR coordinators targets. The same publicity that brought donations could bring RSF attention.
The ERRs navigated this tension carefully, accepting visibility while maintaining operational security. The fixed point required both: public support to sustain resources, covert operation to survive.
Maryam had become part of the nafeer network.
After emerging from hiding on October 27, she had found her way to an ERR distribution point. They had given her food, water, information about escape routes. She had given them her skills—medical assessment, triage, treatment with whatever supplies remained.
“I was a nurse,” she explained. “I’m still a nurse. The hospital is gone, but the practice continues.”
She worked in the field now, moving with displaced populations, providing care in conditions that violated every standard she had been trained to maintain. No sterile environment. No diagnostic equipment. No reliable pharmacy. Only her knowledge, her hands, and whatever materials could be scavenged.
The nafeer network supported her work. Supplies arrived through channels she didn’t fully understand. Volunteers carried her equipment. Information about disease outbreaks and injury clusters reached her through the same networks that distributed food.
“I’m part of something now,” she said. “Not an institution—institutions are destroyed. Something more flexible. More resilient. Something that continues because people choose to continue it.”
The choice was the fixed point’s foundation. Nafeer was voluntary. Every volunteer could stop; every kitchen could close; every network node could go silent. The fixed point held because enough people continued choosing to participate.
The nafeer fixed point represented a different kind of stability.
The other fixed points—international inaction, gold-funded genocide, systematic famine, futile documentation—were stable because no one could break them. They persisted through the failure of countervailing forces.
The nafeer fixed point was stable because people sustained it. It persisted through the success of human choice, the continuous decision to help despite every reason not to.
This was fragile stability—dependent on will, on resources, on the survival of the volunteers. But it was also robust stability—no external force could easily destroy what existed only in the distributed practice of millions.
“The genocide is a machine,” Ibrahim observed. “It operates mechanically, following rules, processing populations. Nafeer is not a machine. It’s a decision, renewed every day by everyone who participates. Machines can be broken. Decisions can only be unmade by those who make them.”
The distinction was the ground of hope. The genocide could continue indefinitely, its fixed point maintained by structural forces. But so could nafeer. The two systems would compete, neither able to eliminate the other, stability against stability.
The war would end someday.
Wars ended. Even this war, this genocide, this catastrophe would eventually reach conclusion. The question was what would remain when it ended.
The genocide would leave bodies, mass graves, traumatized survivors. The international inaction would leave precedent for future inaction. The gold economy would leave destruction and poverty.
Nafeer would leave networks. Communities rebuilt through mutual aid. Skills developed through improvised response. Trust established through reciprocal help.
“When this ends,” Samira said, “we will have practiced solidarity for years. Millions of us. That practice doesn’t disappear. It becomes the foundation for whatever comes next.”
The fixed point of nafeer was, in this sense, an investment in the future. Each day of operation, each kitchen run, each volunteer hour contributed to a structure that would outlast the crisis.
The genocide sought to eliminate. Nafeer sought to persist. Both were succeeding at their objectives. The outcome would be determined not by which succeeded but by the balance between them.
The fixed point held. The kitchens continued. The loop cycled: need → help → trust → need.
The other fixed points were not the whole story.
Definition: Perturbation analysis examines how a system responds to disturbance. A fixed point is stable if small perturbations decay—if the system returns to its equilibrium after being pushed. A fixed point is unstable if small perturbations grow—if any disturbance triggers transition to a different state. Understanding perturbation dynamics reveals what might break the fixed points.
The systems theorist’s question: what would it take?
The fixed points had held for over two years. International inaction, gold-funded genocide, systematic famine, futile documentation—each stable configuration persisted despite pressure, despite protest, despite the accumulated weight of evidence and advocacy.
What perturbation would be sufficient to shift these equilibria?
Leila posed this question formally, applying the analytical training from her professional work. Fixed points could be broken by perturbations that exceeded the system’s stabilizing capacity. The question was: where were the thresholds?
She identified categories of potential perturbation:
Exogenous shock: something outside the system—a natural disaster, an unrelated geopolitical crisis—that reshuffled the variables.
Internal collapse: one of the fixed points breaking down due to its own contradictions, cascading into disruption of the others.
Deliberate intervention: a powerful actor deciding to impose change, accepting the costs necessary to overwhelm the stabilizing forces.
Accumulated pressure: gradual erosion that eventually triggered phase transition, the straw that broke the camel’s back.
Each category had different characteristics. Each implied different strategies for those seeking change.
Exogenous shocks had occurred without breaking the fixed points.
The COVID pandemic had struck Sudan in 2020-2021, straining health systems and economic capacity. The fixed points had survived.
Regional conflicts—Ethiopia’s civil war, instability in Chad—had created spillover effects. The fixed points had absorbed them.
Economic crises in the global gold market, fluctuations in currency values, changes in international commodity flows—all had perturbed the system without triggering transition.
The fixed points were robust to the kinds of shocks that had actually occurred. They had been tested and had held.
“The system has been perturbed repeatedly,” Leila noted. “And each time it returns to the same configuration. This suggests the basin of attraction is large—the fixed points are deeply stable, not precariously balanced.”
The depth of stability was discouraging. If past shocks hadn’t broken the equilibrium, what future shock would?
Internal collapse remained possible.
Each fixed point depended on internal consistency. The gold fixed point required the supply chain to function. The international fixed point required the great powers to remain aligned in inaction. The famine fixed point required humanitarian funding to remain inadequate.
If any of these internal conditions changed, the corresponding fixed point might destabilize.
Consider the gold fixed point. What would happen if the mines were exhausted? If Dubai changed its policies? If internal RSF conflicts disrupted extraction?
Jebel Amer’s gold reserves were substantial but finite. Geological estimates suggested decades of viable extraction at current rates—too long to expect resource exhaustion to break the equilibrium during the war.
Dubai’s policies were driven by profit. As long as Sudanese gold was profitable, Dubai would continue accepting it. Policy change would require external pressure that had not materialized.
Internal RSF conflicts were possible. Hemedti’s control was not absolute; subordinate commanders had their own interests; the factional dynamics of any armed movement created potential for fracture. But so far, the gold had been sufficient to maintain cohesion. When everyone benefited from the supply chain, no one had incentive to disrupt it.
Deliberate intervention was the most direct path but the least likely.
A military intervention—UN peacekeepers with robust mandate, regional force with capacity to impose order, unilateral action by a major power—could break all the fixed points simultaneously. Force could stop the genocide, disrupt the gold trade, enable humanitarian access.
But intervention required will that did not exist.
The US had the capacity but not the interest. Sudan was not strategically vital; intervention would be costly; domestic politics did not favor African military engagement.
The European powers had similar constraints. Their militaries were oriented toward the eastern flank, toward containing Russia. Sudan was peripheral.
The regional powers—Egypt, Saudi Arabia, UAE—had mixed interests. Some supported the SAF; some supported the RSF; none wanted to expend resources on intervention that might fail.
“The capacity exists,” Yusuf had observed during the siege. “The weapons exist. The soldiers exist. The logistics exist. What doesn’t exist is the decision to use them. And decisions don’t emerge from military analysis. They emerge from political calculation that we don’t control.”
The perturbation of intervention was possible but not probable. Analysis could not make it more probable. Only political change in the intervening countries could create the conditions for intervention.
Accumulated pressure was the most common theory of change.
Advocates for Sudan operated on this theory: if enough reports were published, if enough protests occurred, if enough diplomatic representations were made, eventually the pressure would build to the point where the fixed points shifted.
The theory had historical support. Some conflicts had been affected by accumulated advocacy. Some perpetrators had eventually faced justice. Some international interventions had eventually occurred.
But the theory also had historical counter-evidence. Some conflicts had persisted for decades despite continuous advocacy. Some perpetrators had died unpunished. Some populations had been eliminated while the international community watched and documented.
“The problem with accumulated pressure,” Leila said, “is that we don’t know where the threshold is. We don’t know if a threshold exists. We might be one protest away from breakthrough, or we might push forever with no effect.”
The uncertainty was paralyzing. Should advocates continue pushing, hoping they were near the threshold? Or should they conclude that the threshold was unreachable and redirect their efforts?
There was no way to know. The system’s response to future pressure could not be predicted from its response to past pressure.
Ibrahim analyzed the question from his position in the camps.
“We’re asking the wrong question,” he said during a discussion with other activists. “We’re asking what would change the international system. We should be asking what we can change regardless of the international system.”
The reframing pointed to a different kind of perturbation: changes that affected the local dynamics without depending on external intervention.
The ERR networks were such a change. They had not broken the genocide fixed point, but they had created a competing fixed point. The nafeer system operated regardless of international action.
The documentation efforts were similar. They might not trigger intervention, but they created conditions for future accountability. The evidence would exist even if current decision-makers ignored it.
“We can’t make them intervene,” Ibrahim concluded. “But we can ensure that when this ends—whenever it ends, however it ends—there is a record, there are networks, there is a foundation for reconstruction.”
The perturbation was temporal rather than immediate: affecting what would exist after the fixed points eventually broke, rather than breaking them now.
The SAF’s March 2025 recapture of Khartoum was a perturbation.
For months, the RSF had controlled Sudan’s capital. Then, in a series of coordinated offensives, the SAF pushed them out. The city changed hands—or more precisely, changed from RSF control to contested status.
The perturbation had rippled through the system.
The RSF’s aura of invincibility was damaged. The assumption that they would inevitably win was challenged. The populations that had accommodated RSF control began to reconsider their options.
The international community adjusted its calculations. A negotiated settlement became more plausible if the SAF could maintain military pressure. Diplomatic efforts intensified.
The gold supply chains were affected. The SAF’s territorial gains threatened some transit routes. The RSF had to divert resources to military operations that the gold trade’s smooth functioning had previously subsidized.
But the perturbation did not break the fixed points.
The RSF retreated from Khartoum but retained control of Darfur. The gold mines remained under their control. The genocide continued—indeed, accelerated, as the RSF concentrated its ethnic cleansing in the territories it still held.
The international response remained inadequate. More diplomatic activity did not translate into more effective action.
The famine intensified rather than abating, as the military operations disrupted the food systems further.
The Khartoum perturbation was a test case.
A significant military shift had occurred. The system had responded. The fixed points had not broken.
The lesson was sobering. Even substantial perturbations—the recapture of a capital city—were insufficient to trigger phase transition. The fixed points were more stable than advocates had hoped.
“We thought military success would change things,” a SAF officer told Yusuf after the El Fasher escape. “We took Khartoum. We proved we could fight. But the world still doesn’t care. The gold still flows. Our people still starve.”
The perturbation had been absorbed. The system had returned to its equilibrium—slightly different in detail, identical in structure.
What perturbation would be sufficient?
The analysis suggested that only very large disruptions had any chance of breaking the fixed points. The threshold was high—perhaps higher than any realistic perturbation could reach.
Some possibilities:
Complete RSF military collapse—not plausible given their resources and external support.
Total gold supply chain disruption—not achievable without major power intervention in Dubai, which required political will that didn’t exist.
Massive humanitarian intervention—not forthcoming from the international system as currently configured.
Internal RSF fragmentation—possible but not predictable or controllable.
Catastrophic exogenous shock—possible but random, unplannable.
Each potential perturbation was either too small to break the fixed points or too unlikely to serve as a strategic basis for action.
The perturbation analysis yielded a depressing conclusion.
The fixed points were probably too stable to be broken by any intervention that current actors could plausibly execute. The genocide would continue until something changed the fundamental configuration—until one of the internal contradictions accumulated to breaking point, until some unforeseeable shock restructured the system.
This was not hopelessness. Unforeseeable shocks did occur. Internal contradictions did accumulate. Systems that seemed permanent did collapse, sometimes suddenly.
But it was acknowledgment that agency was limited. The activists, the documenters, the ERR volunteers, the diaspora advocates—they could not force the transition. They could only prepare for it, maintain pressure, and ensure that when the transition came, they were ready to influence what followed.
“We’re waiting for an earthquake,” Leila said. “We don’t know when it will come. We don’t know if it will come. But we can prepare for what we’ll do after it comes.”
The preparation was itself meaningful. The networks built, the evidence gathered, the relationships established—these would matter when the fixed points finally shifted.
The perturbation analysis was complete. The next chapter would examine the one perturbation that had actually occurred—the recapture of Khartoum—and what it revealed about the system’s dynamics.
Definition: A case study in perturbation: the SAF’s recapture of Khartoum in March 2025. The event tested the fixed points’ stability, revealing which were robust and which were fragile. The perturbation’s absorption into the system’s dynamics demonstrated the depth of the equilibrium.
The offensive began on January 15, 2025.
After twenty months of war, twenty months of RSF dominance in the capital, the SAF launched coordinated attacks on multiple fronts. The strategy was different from their earlier, failed operations: this time, they had rebuilt capacity, secured supply lines, and coordinated across commands.
The RSF was not expecting the intensity. Their control of Khartoum had become comfortable—checkpoints maintained, civilians suppressed, gold revenues managed. They had spread their forces across the country, pursuing campaigns in Darfur and Kordofan.
The SAF exploited this dispersion. They struck simultaneously at Omdurman, at Bahri, at the airport they had lost nineteen months before. The attacks were not immediately successful, but they forced the RSF to concentrate.
By February, the front lines were moving. By March, the RSF was evacuating central Khartoum. By April, the city was contested but no longer controlled.
Maryam heard the news in El Fasher, where she was still working in the skeleton of South Hospital. The report came through fragmentary communications—a WhatsApp message, a radio broadcast, a rumor that spread through the camp.
“They retook Khartoum?”
“Parts of it. The city center. The RSF still holds the south. But it’s changing.”
The perturbation had arrived.
The immediate effect was psychological.
For twenty months, the RSF had seemed unstoppable. They had taken the capital. They had besieged the remaining SAF strongholds. They had massacred populations without consequence. The momentum had been entirely theirs.
The Khartoum recapture broke that narrative.
In the camps, the refugees began to speak differently about the future. “Maybe we can go home.” “Maybe the war will end.” “Maybe they can be beaten.”
The hope was fragile—a capital recaptured did not mean the war was over—but it existed. The fixed point of RSF inevitability had cracked.
Yusuf, stumbling through the desert toward Chad, heard the news from a herder who had a radio. He stopped walking for an hour, sitting in the shade of an acacia tree, processing the information.
The garrison he had commanded for 920 days had fallen. El Fasher had been lost. But Khartoum had been retaken. The war was not over. The SAF was not defeated.
He continued walking. The destination had not changed—Chad, safety, survival. But the meaning of survival had shifted. He was not fleeing a lost war. He was escaping one front of an ongoing war.
The international response was revealing.
Diplomatic activity increased. The US envoy to Sudan intensified shuttle diplomacy. The UAE began hedging its bets, maintaining RSF support while opening channels to the SAF. The African Union proposed new peace talks.
But the increase in activity did not produce substantive change.
The peace talks yielded nothing. The RSF was not interested in negotiating from weakness. The SAF was not interested in stopping when it was finally winning.
The sanctions remained inadequate. The gold still flowed through Dubai. The weapons still reached RSF forces.
The humanitarian access did not improve. Both sides blocked aid to areas controlled by the other. The famine continued.
“They’re more active now,” Leila observed. “More meetings. More statements. But activity isn’t action. The activity is displacement—doing something to avoid doing something effective.”
The international fixed point had bent but not broken. The perturbation had produced motion without movement.
The gold fixed point was stressed but not disrupted.
The RSF’s loss of Khartoum affected their logistics. Some banking channels were interrupted. Some transit routes required rerouting. The overhead of extraction increased.
But Jebel Amer remained under RSF control. The mines continued operating. The gold continued flowing—perhaps at reduced volume, perhaps through more circuitous routes, but flowing nonetheless.
The Dubai refineries continued accepting Sudanese gold. The UAE had not changed its policies. The profit motive remained aligned with continued trade.
“The gold is resilient,” an analyst explained. “The RSF built redundancy into their supply chains. They have multiple routes, multiple brokers, multiple ways to move product. Losing Khartoum hurts, but it doesn’t break them.”
The fixed point had absorbed the perturbation. The economic substrate of genocide remained intact.
The famine fixed point actually intensified.
The fighting around Khartoum disrupted the agricultural areas south of the city. The population movements created new humanitarian needs. The blockades increased as both sides tried to starve each other out.
By May 2025, the number of people in famine conditions had grown from twenty-five million to thirty million. The perturbation that had briefly seemed hopeful had actually worsened the humanitarian situation.
“War is always worse than non-war,” Samira observed. “The fighting was bad when the RSF won. It’s worse now that the fighting is active. People die in crossfire. People lose access to food when the front line crosses their village. Victory is as destructive as defeat, for the civilians caught between.”
The ERR networks in Khartoum had to relocate repeatedly. Areas they had served became combat zones. Kitchens that had operated for months were destroyed or abandoned. The nafeer network adapted but suffered.
The genocide accelerated in Darfur.
With their attention required in Khartoum, with their aura of invincibility damaged, the RSF intensified their operations in the areas they still controlled. If they could not hold the capital, they would at least complete their project in Darfur.
El Fasher fell in October 2025—seven months after the Khartoum perturbation began. The timing was connected: the RSF, needing a victory to restore their narrative, committed resources to the siege’s conclusion.
The massacre that followed was accelerated, concentrated, almost desperate. The executioners worked quickly, as if they feared interruption.
“They’re rushing,” Maryam observed from her hiding place. “The normal operations take time. They’re not taking time. They’re killing as fast as they can.”
The perturbation in Khartoum had not saved El Fasher. It had condemned El Fasher, by making the RSF more determined to complete their work before the opportunity closed.
Yusuf reached Chad on November 3, 2025.
He crossed the border at a remote point, avoiding the formal checkpoints where he might be identified. The Chadian military found him three days later—a Sudanese officer, SAF uniform fragments still visible under the civilian clothes he had acquired, in desperate condition.
They interrogated him briefly, then sent him to a camp for debriefing. The international observers there wanted his testimony. The SAF command wanted his assessment.
“What happened in El Fasher?”
He told them. The siege. The supply situation. The final assault. The stadium. The escape.
“What’s the situation now?”
“The city is destroyed. The garrison is destroyed. The population...” He paused. “I don’t know how many survived. Thousands died in the stadium. Thousands more in the city itself. Those who escaped—some reached the camps. Most didn’t.”
“Can El Fasher be retaken?”
“There’s nothing to retake. The RSF has completed their operation. The Masalit population of El Fasher is gone. They could retake the land. The people are already dead.”
The perturbation that had recaptured Khartoum had arrived too late for El Fasher. The genocide had outrun the military dynamics.
The perturbation revealed the fixed points’ hierarchy.
The international fixed point had proven most robust—unchanged in substance despite increased activity.
The gold fixed point had proven resilient—stressed but functional, adapting to new circumstances.
The famine fixed point had intensified—growing worse through the dynamics of renewed conflict.
The genocide fixed point had accelerated—responding to the perturbation not by slowing but by speeding up.
Only the nafeer fixed point had shown flexible resilience—adapting, relocating, continuing through changed circumstances.
The ranking was instructive. The positive fixed point was not the most stable, but it was the most adaptive. The negative fixed points were deeply stable but brittle in their response to change—either unaffected or made worse by perturbation.
The Khartoum perturbation’s ultimate outcome was unclear.
The city remained contested as of late 2025. Neither side had achieved decisive victory. The war continued without resolution.
Would the perturbation eventually break the fixed points? The SAF’s survival maintained the possibility. If they could sustain military pressure, if the RSF’s resources depleted, if the contradictions in the RSF coalition intensified—perhaps the equilibrium would eventually shift.
But “eventually” provided no comfort to those dying now. The perturbation had not stopped the genocide, had not fed the starving, had not brought justice to the perpetrators.
“We had hope for a moment,” an ERR coordinator said. “Khartoum fell. It seemed like things might change. Then El Fasher fell. Things got worse. Hope is exhausting.”
The perturbation analysis was complete.
The Khartoum recapture had been the largest disturbance to the system in over a year. Its effects had propagated through all the fixed points. Some had absorbed the perturbation unchanged. Some had shifted in unexpected directions. None had broken.
The conclusion was not that change was impossible—the perturbation had produced change. The conclusion was that change was complex, nonlinear, capable of producing outcomes opposite to what advocates hoped.
Military victory in Khartoum had not ended the war. It had not stopped the genocide. It had not fed the hungry or protected the displaced. It had reconfigured the dynamics without resolving them.
“The system is more robust than we are,” Ibrahim wrote. “We push, it absorbs. We perturb, it stabilizes. The only thing more persistent than the fixed points is our persistence in opposing them.”
The persistence would continue. The fixed points would continue. The competition between stability and change would determine the future.
Part V approached its conclusion.
Definition: Equilibrium restoration is the system’s return to fixed-point stability after perturbation. The disturbance dissipates. The dynamics that maintain the equilibrium reassert themselves. The system, briefly disturbed, settles back into its persistent configuration. Restoration demonstrates the depth of the fixed point’s attracting basin.
By December 2025, the system had stabilized.
The Khartoum perturbation had worked through the dynamics. The initial shock had produced adjustment, the adjustment had produced counter-adjustment, and the system had found a new configuration that preserved its essential structure.
The international community had returned to its baseline of concern without action. The flurry of diplomacy after the SAF’s victories had yielded nothing binding. The statements continued; the inaction continued; the fixed point held.
The gold trade had restructured around the new territorial realities. Some routes had closed; others had opened. The Dubai connection remained intact. The revenue stream had fluctuated but stabilized.
The famine had reached a new plateau of suffering—thirty million hungry, millions in acute crisis—that persisted without triggering interventions sufficient to break it.
The genocide had largely completed its major operations in Darfur. El Fasher had fallen. The Masalit population of West Darfur was killed, displaced, or in camps. The ethnic cleansing had achieved its territorial objectives.
The nafeer networks had adapted to the new configuration—relocated operations, new routes for supplies, adjusted structures for coordination. The positive fixed point persisted alongside the negative ones.
Equilibrium restored.
The restoration was not identical to the pre-perturbation state.
Khartoum remained contested. The SAF’s presence was stronger than it had been in 2024. The war continued in new configurations.
But the essential structure—the fixed points, their relationships, their self-reinforcing dynamics—remained intact. The perturbation had changed details without changing structure.
Leila compared her data from early 2024 to late 2025. The numbers were different. The geographic distribution was different. The specific actors had shifted. But the patterns were the same:
International response: statements without action, activity without effect. Gold flow: continuous, self-sustaining, resilient. Humanitarian situation: chronic crisis, inadequately addressed. Genocide status: ongoing, evolving, incomplete but proceeding. Mutual aid: persistent, adaptive, insufficient to need.
“We’re in the same place,” she concluded. “The GPS coordinates are different, but the systemic coordinates are identical. We’ve moved around the attractor basin without escaping it.”
The perpetrators understood equilibrium restoration intuitively.
The RSF’s response to the Khartoum loss had not been panic. They had retreated strategically, concentrated on Darfur, maintained their economic base. They knew that time was on their side as long as the gold flowed and the international community remained inactive.
“They can take Khartoum,” an RSF commander had said in a leaked communication. “We’ll take it back, or we’ll hold Darfur. Either way, we survive. The international community will eventually accept whatever reality we create. They always do.”
The cynicism was historically informed. The RSF’s predecessors—the Janjaweed—had conducted genocide in 2003-2005. The international community had condemned them, sanctioned them, issued arrest warrants. Twenty years later, the same people ran the RSF, controlled vast territory, traded openly with the world.
The fixed point of impunity was perhaps the deepest of all. Perpetrators could reasonably expect that genocide would not prevent their eventual rehabilitation. The equilibrium included this expectation.
The victims understood equilibrium restoration differently.
For the survivors in the camps, for the displaced scattered across three countries, the restoration of equilibrium meant continuation of suffering. The hope generated by the Khartoum perturbation faded. The expectation of eventual return home receded.
Ibrahim spoke with families who had been in Adré camp for over two years now. Their initial displacement had been emergency—hasty flight, expectation of quick return. Their current situation was something else: permanent impermanence, settled homelessness.
“We’re not refugees anymore,” a woman told him. “Refugees wait to go home. We’re not waiting. There’s nothing to wait for. Our home is gone. We’re something else—I don’t know what word to use.”
The equilibrium had absorbed these people. They were no longer a crisis requiring urgent response; they were a population to be managed, maintained at subsistence, indefinitely.
Maryam emerged from hiding into the restored equilibrium.
The immediate danger had passed. The RSF’s massacre operations in El Fasher had concluded. The city was under their control—destroyed, depopulated, but stable. She was no longer actively hunted; she was merely a survivor in occupied territory.
She found other survivors. Medical personnel who had hidden as she had. Patients who had somehow escaped the hospital killings. Civilians who had emerged from basements and storage rooms and false walls.
They gathered in what remained of a neighborhood that had not been entirely destroyed. They organized—not formally, not with structures, but through the nafeer instinct. Someone had food; they shared it. Someone needed medical care; Maryam provided it. Someone knew a route out; they planned.
“The equilibrium includes us,” she realized. “We’re part of how the system has stabilized. Survivors in the ruins, tolerated as long as we don’t threaten the new order. We’ve been absorbed into the fixed point.”
The absorption was survivable but suffocating. To exist within the equilibrium was to accept its constraints, to limit one’s actions to what the stable configuration permitted.
The ERR networks recalibrated.
The Khartoum operations had expanded during the SAF’s advance. Now, with the city contested but not fully liberated, the networks had to navigate multiple zones of control. Some neighborhoods were SAF territory; some were RSF; some were no-man’s-land.
Samira coordinated across these boundaries. The kitchens served whoever was hungry, regardless of which military controlled their neighborhood. The nafeer primitive did not recognize front lines.
This neutrality was risky. Both sides suspected the ERRs of supporting the other. Both sides interfered with operations. The volunteers had to maintain credibility with multiple armed factions simultaneously.
“We’re threading a needle,” Samira said. “SAF thinks we help RSF civilians, which we do. RSF thinks we help SAF civilians, which we do. Neither trusts us. Both need us. We survive by being useful to everyone.”
The restored equilibrium included this precarious position: mutual aid operating in the gaps between armed powers, tolerated because essential, threatened because independent.
Yusuf began his testimony in December 2025.
The debriefing in Chad had expanded into formal documentation. The ICC investigators wanted his account. The human rights organizations wanted his account. The media wanted his account.
He gave it repeatedly, each telling exhausting, each iteration reinforcing the memories he might otherwise have escaped.
“How many died in the stadium?”
“I don’t know exactly. Thousands. At least three thousand while I was there. More after I escaped.”
“Can you identify any perpetrators?”
“Some commanders. Some fighters I recognized from earlier in the war. Names, I can give you names.”
The names accumulated. They joined Ibrahim’s documentation, joined Maryam’s photographs, joined the satellite imagery and financial records and survivor testimonies. The archive grew.
The archive was part of the restored equilibrium. Documentation continued; prosecution remained impossible. The fixed point of futile witnessing held alongside all the others.
The equilibrium would persist until it didn’t.
This was the fundamental lesson of fixed point analysis: stable configurations persisted until they didn’t. The transition, when it came, might be gradual or sudden, predictable or surprising. But it would come eventually.
Nothing was permanent. The current equilibrium—however deeply stable—was a transient configuration in the longer view of history. The question was not whether it would end but when, and what would follow.
Ibrahim contemplated this in the camp, surrounded by his notebooks, surrounded by survivors whose testimonies he had recorded.
“We’re waiting for history to turn,” he wrote. “The current configuration favors the perpetrators, punishes the victims, exhausts the advocates. But configurations change. They’ve changed before. They’ll change again.”
“Our task is to persist through this configuration. To maintain the networks, the documentation, the hope—however diminished—that makes us ready for the next configuration.”
“The equilibrium is not forever. Nothing is forever. The question is whether we’ll still be here when it ends.”
The children born in the camps would not remember the world before.
There were children now—two years old, three years old—whose entire lives had been lived in displacement. They had no memory of El Geneina, of El Fasher, of homes their parents described but they had never seen.
These children would grow up with the equilibrium as their normal. Famine-level food insecurity would be ordinary. Camps would be home. The genocide would be history that shaped their present without being their experience.
What would they make of the fixed points? Would they accept the equilibrium as natural, as inevitable, as the way things were? Or would they see possibilities their parents had lost sight of?
The restoration of equilibrium created new generations who had never known anything else. The system’s persistence was also its reproduction—each year that passed normalized what had been catastrophic.
Part V concluded with equilibrium restored.
The fixed points had been examined: international inaction, gold-funded genocide, systematic famine, futile documentation, persistent nafeer. Each was stable. Each reinforced the others. Together they formed a compound equilibrium that resisted perturbation.
The Khartoum recapture had tested the system and found it robust. The perturbation had produced adjustment but not transformation. The essential dynamics continued.
This was the reality: a stable configuration of horror, persisting because nothing had broken it, likely to continue until something did.
Part VI would examine universal construction—how both the genocide and the resistance had built complex systems from simple primitives. The construction was ongoing. The systems were still being built.
The equilibrium held. The construction continued. The automaton processed.
The witness witnessed. The record recorded. The help helped.
Nothing had concluded. Everything persisted.
Definition: A universal constructor is a system capable of building arbitrarily complex structures from simple primitives. Given sufficient resources and instructions, the constructor can create anything that can be constructed—including copies of itself. The power of universal construction lies in its generality: the same constructive principles yield bombs and bridges, genocide and survival networks.
John von Neumann defined the problem in 1948.
He was not thinking about genocide. He was thinking about biology—specifically, how living organisms replicated themselves with such high fidelity. How did a cell construct a perfect copy of itself from raw materials? What minimal machinery was required?
His answer was theoretical but precise: a universal constructor needed three components.
Instructions: A description of what to build, stored in some readable format.
A constructor mechanism: Machinery capable of reading instructions and manipulating materials to build what the instructions described.
A copier: A mechanism to replicate the instructions themselves, so the constructed entity also possessed the blueprint for construction.
With these three components, anything could be built that could be described. The constructor was universal because its capabilities were limited only by its instructions, not by its fundamental architecture.
Von Neumann proved that such constructors were logically possible. Biology provided existence proofs: DNA was instructions, cellular machinery was the constructor, DNA replication was the copier. Life itself was universal construction.
The Sudan genocide was universal construction.
The RSF possessed instructions—the accumulated knowledge of how to conduct ethnic cleansing, inherited from the Janjaweed, refined through practice, transmitted through training and example.
The RSF possessed a constructor mechanism—the armed forces, the logistics networks, the command structures capable of translating instructions into physical violence.
The RSF possessed a copier—the processes by which new fighters were trained, new units were formed, the patterns were replicated across expanding operations.
The three components enabled arbitrary-scale construction of genocide. Given sufficient resources—gold—and sufficient operational space—international inaction—the RSF could construct whatever their instructions described: mass graves, emptied territories, eliminated populations.
“They’re not just killing people,” Leila observed during one of her analytical presentations. “They’re constructing a new reality. A Darfur without Masalit. A Sudan without resistance. They’re building something—it just happens that the building requires destroying people.”
The ERR networks were also universal constructors.
They possessed instructions—the nafeer tradition, the organizational patterns developed through the 2019 revolution, the accumulated knowledge of community organizing.
They possessed a constructor mechanism—the volunteer networks, the communication channels, the distribution systems capable of translating need into response.
They possessed a copier—the processes by which new volunteers were trained, new kitchens were established, the patterns were replicated across expanding operations.
The three components enabled arbitrary-scale construction of mutual aid. Given sufficient resources—diaspora support—and sufficient social cohesion—community trust—the ERRs could construct whatever their instructions permitted: kitchens, clinics, evacuation routes, documentation networks.
The same theoretical framework described both systems. Universal construction was morally neutral—a capability that could be directed toward any purpose. The genocide and the resistance were both expressions of universal constructive capacity, differentiated by their instructions rather than their fundamental architecture.
Ibrahim grasped this equivalence as he analyzed his documentation.
“The RSF and the ERRs are mirror images,” he wrote. “Both solve the same computational problem: how do you coordinate millions of people to accomplish a large-scale objective? Both use similar methods: distributed networks, local adaptation, replication of patterns. The difference is the objective—elimination versus preservation.”
The insight was disturbing. It implied that the capacity for genocide was not foreign to the capacity for good. The same organizational principles, the same constructive capabilities, the same human tendencies toward coordination could produce either outcome.
“We like to think evil is different in kind,” Ibrahim continued. “That genocide requires special malevolence beyond normal human range. But it doesn’t. It requires organization. Organization is morally neutral. The malevolence is in the instructions, not the constructor.”
This did not diminish the moral weight of the genocide. The RSF’s instructions were monstrous; the people who followed them bore responsibility; the system they constructed was evil. But understanding the mechanism clarified what was being confronted: not exceptional depravity but organized depravity—the same organizational capacity that built hospitals and schools, directed toward elimination.
Universal constructors had characteristic features.
Modularity: Complex constructions were assembled from simpler components that could be combined in multiple ways. The genocide’s primitives—identify, separate, execute, loot, erase—combined into different operations for different contexts. The ERR’s primitives—identify need, coordinate, deliver, adapt, document—combined similarly.
Scalability: The constructor could operate at any scale, limited only by resources. The RSF could massacre a village or a city using the same basic methods. The ERRs could feed a neighborhood or a region using the same basic methods.
Error correction: The constructor included mechanisms for detecting and correcting failures. When RSF operations encountered resistance, command structures adapted. When ERR distributions failed, volunteers adjusted.
Self-replication: The constructor could build copies of itself. The RSF trained new fighters who trained more fighters. The ERRs trained new volunteers who trained more volunteers.
These features made universal constructors powerful and persistent. Once established, they were difficult to eliminate because they could regenerate from fragments. Kill some RSF fighters; more were trained. Close some ERR kitchens; more opened.
The competition between constructors was a war of building.
The RSF constructed elimination. Each day, their system processed more territory, more people, more of the reality they wanted to create. The construction was additive even when destructive: each mass grave was a construction; each emptied village was a construction; the erasure itself was something being built.
The ERRs constructed survival. Each day, their system preserved more people, more communities, more of the social fabric the genocide sought to destroy. The construction was regenerative: each kitchen was a construction; each evacuation route was a construction; the persistence itself was something being built.
Neither constructor could easily destroy the other. The RSF could kill ERR volunteers, but the copier mechanism produced more. The ERRs could save lives, but the RSF’s constructor produced more deaths. Both systems operated simultaneously, building competing realities.
“The war will be decided by which constructor runs out of resources first,” Leila suggested. “Or by which instructions prove more sustainable. The genocide requires continuous violence. The mutual aid requires continuous solidarity. Which exhausts first?”
The universal constructor framework revealed intervention points.
To disable a constructor, you could target instructions—destroy the knowledge of how to operate. But instructions were distributed, replicated, carried in the minds of many. Destroying instructions was nearly impossible.
You could target the constructor mechanism—eliminate the personnel and equipment that executed instructions. But the copier would regenerate the mechanism. Destroying the mechanism required faster destruction than the copier could regenerate.
You could target resources—cut off the inputs that enabled construction. For the RSF, this meant the gold. For the ERRs, this meant diaspora support and local supplies. Resource deprivation was the most effective intervention because it limited all constructive capacity simultaneously.
“The international community’s refusal to target the gold trade is, in effect, a decision to let the genocide constructor continue operating,” Leila argued. “They could disrupt the resources. They’ve chosen not to. That’s not neutrality; that’s enablement.”
The argument was rejected by those who preferred the language of neutrality. But the universal constructor framework made the choice stark: either you disrupted the genocide constructor’s resources, or you permitted its continued operation.
Maryam experienced herself as a component in a constructor.
She had not chosen to be part of a system. She had chosen to be a nurse, to heal, to help. But her individual choices aggregated into a larger pattern. She was a node in the ERR network, executing primitives—medical assessment, treatment, documentation—that contributed to the network’s overall construction.
“I’m a machine,” she said to a colleague one night, not bitterly but observationally. “A component in a machine. The machine produces survival, which is good. But I’m still a component.”
The colleague disagreed. “You’re not a machine. Machines don’t choose. You chose to be here. You could stop.”
“Could I? Can any of us stop? The need is everywhere. The skills are mine. What choice do I really have?”
The question probed the boundary between component and agent. The universal constructor framework treated individuals as nodes, as carriers of primitives, as mechanisms for instruction execution. But individuals experienced themselves as choosing, as deliberating, as morally responsible.
Both perspectives were valid. Maryam was a component in a system; she was also a person making choices. The system emerged from choices; the choices were shaped by the system. The distinction between constructor and constructed was not absolute.
Von Neumann had also theorized about self-reproducing automata.
A constructor that could build copies of itself could, in principle, expand without limit. Each copy could make more copies. The population would grow exponentially until resource constraints limited it.
Both the RSF and the ERR networks exhibited self-reproduction. The RSF trained fighters who trained more fighters. The ERRs trained volunteers who trained more volunteers. Both were self-reproducing constructors, expanding their capacity through replication.
The RSF’s replication was limited by recruitment—not everyone would become a killer. The ERR’s replication was limited by resources—volunteers still needed food, supplies, operational support.
The different limiting factors produced different dynamics. The RSF could expand rapidly when recruitment succeeded and gold flowed. The ERRs could expand rapidly when solidarity was strong and diaspora support arrived.
In the current configuration, both constructors were approximately in steady state—reproducing themselves, maintaining their capacity, neither expanding dramatically nor collapsing. The competition was between two stable constructors, not between one growing and one shrinking.
The universal constructor concept unified the analysis.
The genocide and the resistance were both expressions of humanity’s capacity to build complex systems from simple primitives. The same fundamental capability produced death and survival, elimination and preservation.
This was the von Neumann perspective: not moral judgment but mechanical analysis. How did the systems work? What were their components? How did they replicate? What resources did they require?
The analysis did not replace moral judgment—the genocide remained evil; the resistance remained good. But it supplemented moral judgment with understanding. To oppose the genocide effectively required understanding how it operated, what made it persistent, where its vulnerabilities lay.
The universal constructor was an analytical tool, not a moral framework. It revealed structure without assigning value. The values came from elsewhere—from human choice, from ethical commitment, from the irreducible distinction between building life and building death.
Part VI would continue examining universal construction in specific domains: the RSF’s construction, the ERR’s construction, the diaspora’s construction, the documentation’s construction. Each domain instantiated the same principles with different instructions, producing different outputs.
The automaton was a constructor. The constructor was building.
What it built depended on its instructions.
Definition: The RSF as universal constructor: a system that builds genocide from primitives. The construction has three outputs: territorial control, ethnic elimination, and economic extraction. Each output is constructed using the same methods—the same primitives, the same rules, the same replication. Understanding the RSF as constructor reveals its systematic nature.
The instructions were inherited and refined.
The Janjaweed had developed templates in 2003: how to identify target populations, how to separate them, how to execute and erase. These templates were not written in manuals—they existed as shared knowledge, transmitted through training and example.
When the RSF formed in 2013, it inherited these templates. The institutional transformation changed names and structures but preserved operational knowledge. The fighters who became RSF brought their expertise with them.
The templates evolved through application. Each operation provided feedback. Methods that worked were replicated; methods that failed were abandoned. By 2023, the RSF possessed highly refined instructions for large-scale ethnic cleansing.
“They’ve optimized,” a military analyst observed. “Twenty years of practice. They know exactly what works. The El Fasher operation wasn’t improvised—it was the execution of a perfected methodology.”
The constructor mechanism was the RSF’s organizational structure.
At the top: Hemedti and the senior commanders, providing strategic direction and resource allocation.
Below: regional commanders, translating strategy into operational planning for their areas.
Below: unit commanders, executing operations according to plans, adapting to local conditions.
At the base: individual fighters, carrying out primitives—identifying, separating, executing, looting, erasing—as components in larger operations.
The structure was hierarchical for command but networked for execution. Units could operate independently when necessary, following the templates without requiring continuous direction. The distributed architecture made the system resilient to decapitation—losing any individual component did not disable the whole.
Bashir was a node in this structure. He received instructions—which village, which population, which operation. He executed according to templates he had internalized over twenty years. He did not need to understand the strategic picture; he needed only to perform his function.
The copier mechanism was training and recruitment.
New fighters were recruited from tribal networks, from displaced populations, from the economically desperate. The gold-funded wages provided incentive. The tribal affiliations provided social reinforcement.
Once recruited, fighters were trained—not in formal academies but through operational integration. They accompanied experienced fighters on operations. They learned the templates through observation and practice. They became carriers of the instructions through embodied experience.
The training process had high fidelity. New fighters learned the same methods their trainers had learned. The templates replicated with minimal variation. A fighter trained in 2023 used essentially the same methods as a fighter trained in 2003.
“They’re transmitting a culture,” a researcher said. “Not just skills—a whole way of seeing the world. The targets are subhuman. The operations are justified. The rewards are deserved. The training doesn’t just teach them how to kill; it teaches them why killing is acceptable.”
The cultural transmission was the copier’s deepest layer. It reproduced not just behavior but belief.
The constructed output was a transformed territory.
In West Darfur, the RSF had constructed a new demographic reality. The Masalit population was gone—killed, displaced, scattered. The land they had occupied was empty or newly settled by Arab communities the RSF favored.
The transformation was physical: villages destroyed, infrastructure demolished, evidence erased.
The transformation was social: communities shattered, families separated, networks of relationship severed.
The transformation was political: local governance eliminated, RSF authority established, resistance crushed.
The construction was comprehensive. Every aspect of the previous reality was dismantled and replaced. The RSF built not merely through destruction but through reconstruction—establishing new facts on the ground that would persist after active operations ended.
The economic extraction was integrated into the construction.
Each operation extracted wealth: cattle from the villages, gold from the mines, property from the displaced. The extraction was not incidental to the genocide; it was co-constructed.
The gold mines required control of territory. The genocide provided that control. The gold funded the genocide. The genocide enabled more extraction. The loop was a construction in itself—an economic machine built from the same primitives as the ethnic cleansing.
“They’re building an economy,” Leila said. “Not a legitimate economy—a predatory economy. But it has structure. It has logic. It produces outputs. It’s a construction like any other.”
The economic construction would persist even if the active genocide concluded. The mines would still be controlled. The extraction would still operate. The wealth would still flow. The RSF had built something durable.
The RSF’s construction had limits.
The constructor could operate only where resources permitted. When ammunition ran low, operations slowed. When fuel was scarce, mobility decreased. The gold sustained the construction, but gold was finite, and conversion to operational capacity was not instantaneous.
The constructor required willing carriers. Not everyone could be made into a killer. The recruitment pool was large but not infinite. Some fighters defected. Some refused orders. The human substrate was imperfect.
The constructor faced opposition. The SAF resisted. The ERR networks provided alternatives. The documentation created records that might eventually enable accountability. The construction was contested, not unopposed.
These limits constrained but did not prevent construction. The RSF adapted to limits—prioritizing operations where resources were sufficient, recruiting where willing populations existed, targeting opposition where it was vulnerabe. The constructor was flexible within its constraints.
Yusuf had watched the construction from inside its target.
As garrison commander at El Fasher, he had observed the RSF’s methods with professional analysis. The siege was textbook: establish encirclement, control supply, apply pressure, wait for collapse. The execution after the fall was similarly methodical.
“They’re good at this,” he admitted to himself, not as praise but as acknowledgment. “They’ve solved the organizational problems. They know how to move forces, how to sustain operations, how to achieve objectives. If I were judging them purely as a military organization, I would have to give them high marks.”
The professional competence made the moral horror worse. These were not chaotic mobs; they were organized, trained, disciplined forces applying refined methods. The genocide was not passion; it was procedure.
“That’s what makes it genocide,” Yusuf concluded. “Passion burns out. Procedure continues until completion.”
The RSF’s construction was self-documenting.
They maintained records—not for historical accountability, but for operational tracking. Which areas had been cleared. Which populations had been processed. Which resources had been extracted.
These records were evidence. When they leaked or were captured, they revealed the systematic nature of the construction. The bureaucratic documentation proved intent and scale in ways that survivor testimony alone could not.
“They’re writing their own indictment,” an ICC investigator observed. “Every form they fill out, every report they file—it’s evidence. They’re so confident of impunity that they don’t even try to hide what they’re doing.”
The confidence was historically grounded. The 2003 genocide had been documented, its perpetrators identified, its records preserved. Nothing had happened. The RSF reasonably expected the same outcome.
The RSF’s construction was ongoing.
As of late 2025, the major operations in Darfur were largely complete. El Fasher had fallen. The Masalit population of West Darfur was eliminated as a coherent presence. The territorial objectives had been achieved.
But construction continued in other forms. The consolidation of control. The establishment of administration. The normalization of the new reality. The prevention of return.
The constructor was transitioning from active elimination to maintenance of accomplished facts. The same organizational capacity that had built the genocide now built the post-genocide order—policing, governance, economic management.
“The construction doesn’t end when the killing ends,” Leila observed. “It continues into the future. They’re building the reality that will persist for decades. The massacres are foundation; what follows is superstructure.”
The universal constructor analysis revealed the RSF’s vulnerability.
The system depended on resources, particularly gold. Disrupting the gold flow would constrain all constructive capacity—military operations, recruitment, equipment maintenance, everything.
The system depended on impunity. If perpetrators began facing consequences—arrests, asset seizures, travel bans with teeth—the calculations that sustained participation might shift.
The system depended on the templates’ acceptance. If the fighters began to reject the instructions—if defection became common, if refusal spread—the constructor would malfunction.
Each vulnerability was a potential intervention point. None had been effectively targeted. The international community had not disrupted the gold. The accountability mechanisms had not imposed consequences. The counter-narratives had not undermined acceptance.
The constructor continued operating because its requirements continued to be met. Breaking the constructor required breaking one of its requirements. Nothing had broken yet.
The RSF as universal constructor: efficient, adaptable, well-resourced, confident of impunity. A machine for building genocide from primitives.
The analysis was complete. The next chapter would examine the counter-constructor: the ERR networks building survival from the same fundamental capacity.
Definition: The ERR as universal constructor: a system that builds survival from primitives. The construction has three outputs: immediate aid, community preservation, and future capacity. Each output is constructed using counter-methods—counter-primitives, counter-rules, counter-replication. The ERR constructor opposes the genocide constructor without mirroring it.
The instructions were cultural and adaptive.
Unlike the RSF’s inherited templates of violence, the ERR’s instructions emerged from nafeer tradition combined with modern organizational innovation. The 2019 revolution had produced Resistance Committees—neighborhood structures for protest coordination. When war came, these structures repurposed.
The instructions were less rigid than the RSF’s. They were principles rather than procedures: help those in need, coordinate through communication, adapt to circumstances, document what happens. The flexibility was intentional—fixed templates couldn’t respond to the chaos of war.
“We don’t have a manual,” Samira explained. “We have values. The values guide adaptation. Every situation is different, but the values stay constant: everyone deserves help, no one gets left behind, we’re in this together.”
The values-based instructions enabled rapid adaptation while maintaining coherent identity. New volunteers didn’t need extensive training in procedures; they needed internalization of principles. Once internalized, the principles generated appropriate behavior in novel situations.
The constructor mechanism was the volunteer network.
No hierarchy commanded the ERRs. Coordination was horizontal—networks of individuals and groups connected through communication channels, sharing information, distributing resources, supporting each other’s operations.
The structure emerged rather than being designed. When war began, people who knew each other started helping. The people they helped joined the helping. The networks grew through organic accretion, not top-down organization.
Aisha was a node in this structure—not because anyone appointed her but because she had developed connections, had proven reliable, had become someone others turned to. Her authority was informal, based on demonstrated competence and trust.
“No one’s in charge,” she said. “Which is the point. If someone was in charge, you could capture them, kill them, and the network would collapse. With no one in charge, there’s no center to attack.”
The decentralization was both principle and protection. It made the ERRs resilient to decapitation attacks that would destroy hierarchical organizations.
The copier mechanism was example and relationship.
New volunteers joined by seeing existing volunteers. They watched, participated, learned through doing. The templates they acquired were not documented procedures but embodied practices—how to run a kitchen, how to coordinate distribution, how to navigate checkpoints.
The transmission was personal. Each volunteer learned from specific people, in specific contexts, through specific experiences. The instructions they internalized bore the marks of their teachers, their local conditions, their particular histories.
This produced variation. The ERRs in Khartoum operated differently than the ERRs in camps in Chad. The methods adapted to contexts while maintaining the underlying values. The copier was high-fidelity for principles, flexible for implementations.
“We’re all ERR, but we’re all different,” a coordinator observed. “The kitchen in Bahri is not the same as the kitchen in Adré. They serve the same purpose, they share the same values, but they do things their own way.”
The variation was a feature, not a bug. It enabled local optimization while preserving systemic identity.
The constructed output was survival itself.
Each meal served was a construction. Each evacuation completed was a construction. Each medical intervention was a construction. Each family kept alive was a construction.
The cumulative output was massive: millions of people assisted over two years. The numbers were not counted precisely—the decentralized structure made comprehensive tracking impossible—but the scale was evident from the scope of operations.
Beyond immediate survival, the ERRs constructed something less tangible: community. The shared experience of giving and receiving help built relationships. The networks formed through nafeer became bonds that would persist beyond the immediate crisis.
“We’re building a society,” Samira said. “Not deliberately—we’re just helping people. But the helping creates connections. The connections become community. When this ends, we’ll have something: people who know each other, who’ve worked together, who trust each other. That’s a construction.”
The ERR construction was inherently regenerative.
Unlike the RSF’s construction, which extracted and depleted, the ERR construction replenished. Help given created capacity for more help. Trust built enabled more coordination. Skills developed through volunteering multiplied through teaching.
The economics were different. The RSF’s gold was finite—extracted from the ground, consumed in operations. The ERR’s resources were renewable—solidarity produced more solidarity, help produced more helpers.
This renewable quality gave the ERR constructor a different temporal profile. The RSF could construct rapidly but faced eventual depletion. The ERRs constructed more slowly but sustainably. The competition was between a sprint and a marathon.
“They have more firepower,” an analyst observed. “We have more staying power. They can destroy faster than we can rebuild, but we can outlast them. The question is whether we can survive long enough for our advantages to matter.”
The ERR construction faced its own limits.
Resources remained scarce. Diaspora support was substantial but not unlimited. Local supplies were disrupted by the war. The gap between need and capacity was always present.
Volunteer fatigue accumulated. People could not sustain twenty-hour days indefinitely. The psychological toll of constant crisis wore down even the most committed. Burnout removed experienced coordinators from the networks.
Security threats persisted. The RSF targeted ERR operations when they could identify them. Working under such threat required constant caution that drained energy and limited reach.
These limits constrained the ERR construction. The constructor operated below its potential capacity because the resources and conditions did not permit full operation.
Maryam worked as a component in the ERR constructor.
Her medical skills were primitives the constructor assembled into larger operations. She assessed patients, provided treatment, made triage decisions, documented what she encountered. Each action was a small construction; the accumulation was the survival of her patients.
She had not chosen this role abstractly. She had chosen to help, and the system had structured her help into a form that aggregated with others’ help. The constructor emerged from individual choices without anyone designing it.
“I thought I was just being a nurse,” she said. “Now I realize I’m part of something larger. The system exists because people like me do what we do. We create it by our actions, and then it shapes our actions.”
The recursive relationship between individual and system was characteristic of universal constructors. The components built the constructor; the constructor organized the components. Neither existed independently.
The ERR construction was documented differently.
The RSF documented for operational control. The ERRs documented for testimony and accountability. The same primitive—documentation—served opposite purposes in the two systems.
The ERR documentation was fragmentary by nature. Volunteers took photographs when they could. Survivors gave testimonies to whoever would listen. Data accumulated in spreadsheets and WhatsApp groups and personal notebooks.
Leila tried to aggregate this documentation into coherent analysis. The diaspora coordination platform assembled reports from hundreds of sources. But the picture was always incomplete, always partial, always catching up to events that moved faster than documentation could capture.
“We’re building a record while running a race,” she said. “The record matters—it will matter enormously after this ends. But we can’t stop running to perfect the record. So we get what we can, and we accept incompleteness.”
The ERR constructor and the RSF constructor competed for the same substrate: the Sudanese population.
Each person helped by the ERRs was a person potentially preserved from the RSF. Each person killed by the RSF was a person the ERRs couldn’t help. The population was finite; the constructors competed for it.
But the competition was asymmetric. The RSF sought elimination. The ERRs sought preservation. The ERRs’ success was the population’s survival; the RSF’s success was the population’s destruction.
“We’re not fighting them directly,” Samira observed. “We can’t—they have guns, we have spoons. We’re racing them. Can we preserve more than they can destroy? That’s the competition.”
The competition had no referee, no scoreboard, no declared outcome. It proceeded through daily interactions: a family saved here, a family killed there; a community preserved here, a community destroyed there. The aggregate result would be determined by the cumulative balance.
The ERR as universal constructor demonstrated an alternative.
The same fundamental capacity—building complex systems from simple primitives—could serve life instead of death. The choice was in the instructions, not the architecture.
This was hopeful and sobering. Hopeful because it showed that human organizational capacity was not inherently aligned with destruction. Sobering because it showed that choice, not structure, determined outcomes. The ERRs had chosen to build survival; the RSF had chosen to build genocide. Both were possible. Both were human.
“We’re proof that another way is possible,” an ERR volunteer said. “They chose one direction; we chose another. They prove nothing about human nature except that some humans choose evil. We prove nothing except that some humans choose good. The choice is always there.”
The universal constructor framework unified the analysis without collapsing the moral difference. The RSF and the ERRs used the same constructive principles toward opposite ends. Understanding the shared framework illuminated both the horror of the genocide and the heroism of the resistance.
The constructor was a tool. The instructions determined its use. The instructions were chosen by humans.
The choice continued. The construction continued. Both constructors were still building.
Definition: The diaspora constructor builds support from distance. Its primitives are money, information, and influence, transmitted across geographic separation. The diaspora cannot build directly—cannot run kitchens or evacuate families—but can construct the conditions that enable direct builders. The diaspora constructor is a meta-system: a constructor that builds other constructors.
The diaspora was scattered before the war.
Sudanese communities existed in the Gulf states, in Egypt, in Europe, in North America. They had emigrated for education, for work, for safety during previous conflicts. They maintained connections—WhatsApp groups, family visits, remittances—but the connections were individual, fragmented.
The war transformed scattered individuals into a coordinated system.
Within days of April 15, 2023, diaspora groups began organizing. The coordination was emergent: someone created a GoFundMe, someone else shared it, someone translated needs into donation requests. The system self-assembled because the components—people with resources and connection to Sudan—recognized their complementary capacities.
Leila was one node in this assembly. Her technical skills became valuable as the need for data coordination grew. Her position in London gave her access to media and policy networks. Her family connections gave her legitimacy with communities that distrusted outsiders.
“We didn’t plan this,” she observed. “None of us woke up thinking ‘I’ll help build a transnational humanitarian network today.’ But that’s what we built. The need called it into existence.”
The diaspora’s primary primitive was money.
The transfer mechanisms already existed: hawala networks, formal remittance services, informal channels through travelers and traders. The infrastructure for moving value from diaspora to Sudan had operated for decades.
The war redirected these flows. Money that had gone to family support—school fees, medical bills, daily expenses—now went to emergency aid. The same channels carried different content.
The scale was substantial. Leila’s tracking estimated $200 million in the first two years—possibly more, since informal transfers escaped documentation. The money funded ERR operations, paid for supplies, enabled evacuation routes.
“We’re the cash machine,” a diaspora coordinator said bluntly. “That’s our role. The people on the ground do the work. We provide the fuel. Without us, the ERRs would collapse. Without the ERRs, our money would have nowhere to go.”
The interdependence was mutual. The diaspora constructor needed ground-level constructors to translate money into action; the ground-level constructors needed diaspora resources to sustain operations.
The diaspora’s second primitive was information.
The communities in Khartoum and Darfur knew what was happening to them but could not always communicate it. The blackouts, the blockades, the dangers of documentation—all constrained information flow.
The diaspora served as amplifier and router. Messages that reached one diaspora node could be relayed to media, to policymakers, to other nodes. The scattered geography became an advantage: a blackout in Sudan didn’t silence communities in London or Minneapolis.
“We’re the voice,” Leila said. “When they can’t speak, we speak for them. When their testimonies can’t get out, we’re the channel. The geographic distance that separates us also protects us.”
The information function included translation—not just linguistic but contextual. Diaspora members could explain Sudanese realities to foreign audiences, could frame local events in terms that international media understood, could bridge cultural gaps that raw information couldn’t cross.
The diaspora’s third primitive was influence.
Some diaspora members had access to power—positions in government, connections to business, relationships with media figures. These relationships could be leveraged for advocacy.
The influence was limited. No diaspora community had enough political weight to force intervention in Sudan. But influence at margins mattered: a meeting with a congressional staffer, a story placed in a major newspaper, a contact who could push a bureaucratic process forward.
“We’re not powerful,” an advocate acknowledged. “But we’re not nothing. Each small push accumulates. Each contact we make, each story we tell, each official we educate—it adds up. Maybe not to anything decisive. But to something.”
The influence function was frustrating because it produced so little visible result. The advocacy continued; the genocide continued; the connection between advocacy and outcome was unclear. But the function persisted because the alternative—not trying to influence—seemed worse.
The diaspora constructor had characteristic limitations.
Distance created disconnection. Diaspora members knew Sudan through screens and stories, not through lived experience. They could misunderstand situations, could prioritize wrongly, could impose preferences that didn’t match ground realities.
“They don’t know what it’s like,” an ERR coordinator complained to Leila. “They think they know. They send money and give advice. But they’re not here. They don’t feel the fear, smell the death, make the choices we make.”
The critique was fair. The diaspora constructor was powerful but indirect, influential but removed. Its outputs were shaped by distance as much as by intention.
The diaspora also faced coordination problems. Multiple organizations pursued similar goals without consolidation. Donors duplicated some efforts while neglecting others. The emergent nature of the system meant no one optimized the whole.
Leila tried to address this through her coordination platform, but coordination was voluntary, and not everyone participated. The diaspora remained less organized than it could have been.
The diaspora constructor was itself built from smaller constructors.
Each community organization was a constructor: taking inputs (money, volunteers, information), processing them (events, fundraisers, advocacy campaigns), producing outputs (funds transferred, stories told, contacts made).
These smaller constructors coordinated loosely through shared information and occasional collaboration. The larger diaspora constructor emerged from their aggregated activity without any single entity directing the whole.
The emergent architecture was similar to the ERRs: decentralized, adaptive, resilient to disruption. Attacking any single organization wouldn’t disable the diaspora response. The distributed nature was protective.
“We’re an ecosystem,” a coordinator observed. “Not an organization. Organizations can be killed. Ecosystems adapt and continue.”
The diaspora constructor enabled the ground-level constructors.
The relationship was symbiotic. The ERRs operated because the diaspora funded them. The diaspora had purpose because the ERRs existed to support. Neither would function the same without the other.
The enabling relationship created responsibility. When diaspora money was misused, it harmed the system’s credibility. When diaspora information was wrong, it misled ground-level planning. The diaspora’s choices had consequences for people they would never meet.
Leila felt this responsibility acutely. Each data point she collected, each report she compiled, each recommendation she made—all of it might affect decisions that affected lives. The distance that protected her also burdened her.
“I can’t see the results directly,” she said. “A kitchen I’ve never visited serves food to people I’ll never meet, using money I helped route from donors I’ve never seen. The connection is real but invisible. I’m building something I can’t directly observe.”
The diaspora constructor was reshaping the diaspora itself.
The experience of coordinated response—of working together toward shared purpose, of developing skills in organizing and advocacy—was changing the scattered communities. They were becoming more connected, more organized, more politically engaged.
“After this ends, we won’t be the same,” an activist observed. “We’ve learned how to work together. We’ve built networks. We’ve developed capacities. Whatever Sudan looks like when peace comes, the diaspora will be different—more capable, more connected.”
The self-transformation was an unintended output. The diaspora had organized to help Sudan; in organizing, they had helped themselves. The constructor built not only its intended outputs but also new versions of its own components.
The diaspora constructor would outlast the immediate crisis.
When the war eventually ended—however it ended—the diaspora networks would remain. The relationships built, the skills developed, the platforms created—all would persist.
This persistence was itself a construction: the building of future capacity. The diaspora that would engage with post-war Sudan would be different from the diaspora of April 2023. It would be more organized, more experienced, more connected to ground-level realities.
“We’re building for after,” Leila said. “Not just responding to now. Every system we create, every relationship we establish—it’s infrastructure for reconstruction. We’re building the tools we’ll need when the building of peace becomes possible.”
The temporal extension was strategic and hopeful. The present was terrible, but the present was not forever. The diaspora constructor was building not just for survival today but for rebuilding tomorrow.
The diaspora constructor demonstrated the reach of universal construction.
Construction did not require physical presence. It did not require direct action. It could operate across distance, through intermediaries, by enabling other constructors.
The diaspora’s contribution was real even if indirect: every meal served by an ERR kitchen partly resulted from diaspora funding. The construction was distributed across the entire network, not located at any single point.
This distributed construction was characteristic of complex systems. No single node built everything. The outputs emerged from the interaction of many nodes, each contributing their capacity.
The diaspora was one layer in a multi-layer constructor: gathering resources from one context, transforming them through transfer mechanisms, enabling construction in another context. The layers were interdependent; each needed the others.
The universal constructor analysis applied across layers: primitives, mechanisms, copiers, outputs—all present at each level, all interacting to produce the aggregate system.
The construction continued. The diaspora continued supporting. The ERRs continued operating. The layers continued enabling each other.
Building survival from scattered resources across continents.
Definition: The documentation constructor builds evidence from observation. Its output is not physical structures but information structures—archives, databases, testimonies that encode what happened. The documentation constructor serves the future: building a record that might enable accountability when present conditions do not.
Ibrahim’s archive had grown beyond his notebooks.
The seventeen original notebooks remained, but they had become a small part of a larger system. His testimonies connected to Leila’s databases. His photographs linked to Maryam’s medical records. His locations cross-referenced with Yale HRL’s satellite imagery.
The documentation constructor assembled these fragments into something greater than any single source. The construction was synthetic—combining elements from many inputs to produce comprehensive understanding.
“Each source is partial,” Ibrahim explained. “My notebooks have testimonies but no physical evidence. The satellites have imagery but no human stories. The medical records have injuries but no context. Together, they create a picture that none could create alone.”
The synthesis was itself a construction: taking raw documentation primitives and building them into evidentiary structures.
The primitives of documentation were observational.
Testimony: accounts from survivors, witnesses, perpetrators—anyone who experienced or observed events.
Physical evidence: photographs, videos, objects, locations—tangible traces of what occurred.
Records: written documents, communications, transactions—systematic traces in bureaucratic systems.
Remote sensing: satellite imagery, signal intercepts, environmental data—observation from distance.
Each primitive captured different aspects of reality. Testimony captured experience and intention. Physical evidence captured results and locations. Records captured administrative processes. Remote sensing captured patterns invisible from ground level.
The documentation constructor combined these primitives through triangulation—matching testimony to evidence, correlating records with imagery, building confidence through multiple independent sources.
The constructor mechanism was distributed across organizations.
The ICC gathered evidence for prosecution. Amnesty International gathered evidence for advocacy. Yale HRL gathered imagery for academic analysis. SIHA gathered testimonies for community voice. Physicians for Human Rights gathered medical evidence for specialized documentation.
Each organization had particular competencies and particular perspectives. The aggregation of their work was not centrally planned—no one directed the overall documentation effort. But their outputs combined into a collective record.
“We’re all building the same structure from different angles,” an ICC investigator said. “We don’t coordinate because coordination isn’t necessary. We know the others are working. Our pieces will fit together eventually.”
The eventual assembly was itself a construction: future researchers and prosecutors would combine what current documenters created separately.
The copier mechanism was professional training and organizational culture.
Documentation methods were transmitted through institutions. Human rights organizations trained new researchers in interview techniques, evidence preservation, ethical protocols. These methods propagated through organizational learning.
The universities trained the next generation of forensic investigators, journalists, researchers. The methods they learned were built on what previous generations had developed. Each cohort inherited and refined the documentation primitives.
“I learned from people who documented previous genocides,” an investigator said. “They learned from people who documented earlier horrors. It’s a lineage. The methods improve over time. We’re better at this than we used to be.”
The improvement was measurable. Documentation of the Sudan genocide was more comprehensive, more technically sophisticated, more legally robust than documentation of previous events. The constructor had evolved through replication with refinement.
The constructed output was a record for the future.
The archive would outlast the present. When the war ended, when the political configurations shifted, when the conditions for accountability emerged—the documentation would be available.
This temporal construction was an act of faith. The documenters could not know if their work would ever be used. They could only build the record and hope that future circumstances would enable its application.
“We’re writing for readers who don’t exist yet,” Ibrahim said. “Future prosecutors. Future historians. Future Sudanese who want to understand what happened. We don’t know if they’ll exist, but we build the record as if they will.”
The faith was grounded in historical experience. The documentation of other genocides had eventually been used. Trials had happened, sometimes decades later. The record had mattered.
The documentation constructor faced systematic opposition.
The RSF’s erasure rule was specifically designed to counter documentation. Kill witnesses. Burn records. Block access. Impose blackouts. The erasure constructor and the documentation constructor were in direct competition.
The competition was asymmetric in different ways. The RSF had physical control of most territory—they could destroy evidence, prevent access, eliminate documenters. But the documentation constructor had redundancy—copies in multiple locations, sources that couldn’t all be silenced, satellite observation beyond physical reach.
“They can’t kill the satellites,” a Yale researcher noted. “They can’t erase the images we’ve already captured. They can destroy ground-level evidence, but they can’t destroy the record we’re building from space.”
The competition produced an accumulating advantage for documentation. Each day that passed added to the record. Each attempted erasure was itself documented. The archive grew faster than it could be destroyed.
Maryam’s contribution was medical documentation.
The injuries she treated told stories. Gunshot patterns indicated execution. Malnutrition severity indicated siege conditions. Rape injuries indicated systematic sexual violence. The bodies were evidence; her records encoded what the bodies revealed.
She had developed notation systems for efficiency—shorthand that captured injury types, locations, severity. The notation was her own invention, later shared with other medical documenters, becoming a primitive in the larger documentation system.
“Every patient is a case study,” she said. “I treat them, of course—that’s primary. But I also record. The treatment might save them today. The record might serve justice tomorrow.”
The dual function—healing and documenting—was characteristic of the documentation constructor. The same actions that served immediate needs also built the archive. The primitives were multi-purpose.
The documentation constructor had unintended outputs.
The process of documentation affected the documenters. Ibrahim’s continuous engagement with atrocity testimony changed him. Maryam’s steady recording of injuries accumulated psychological weight. The documentation was not cost-free.
“I carry all of it,” Ibrahim said. “Every testimony I’ve recorded. Every name I’ve written down. They don’t leave me. The archive is external, but I’ve become an archive too.”
The internalization was both burden and asset. The documenters understood the events they recorded with intimacy that distant analysts couldn’t achieve. But they also bore trauma that distant analysts didn’t share.
The documentation constructor processed not only events but also the people who documented events. It transformed its components as it operated.
The documentation constructor would eventually complete.
Someday—no one knew when—the active genocide would end. The documentation would transition from present-tense recording to past-tense archiving. The construction phase would conclude; the preservation phase would begin.
The archive would then exist as a completed structure, available for whatever uses the future demanded. Historical analysis. Legal prosecution. Memorial construction. Educational curricula.
“We’re building a time capsule,” an archivist said. “It won’t be opened until later. We don’t know when later is. We just build it as well as we can and trust that it will matter.”
The completion would not be announced. Documentation would simply stop when there was nothing new to document. The construction would end through exhaustion of inputs, not through declaration of victory.
The documentation constructor was the slowest-acting of the survival constructors.
The ERRs built immediate survival—meals today, evacuations this week. The diaspora built medium-term capacity—funds flowing this month, advocacy over these years. The documentation constructor built for the longest term—records for decades, centuries.
The temporal scale meant the documentation constructor’s impact couldn’t be assessed in the present. Its value would emerge only later, when the conditions that prevented justice today had changed.
“We’re planting trees we’ll never see mature,” Ibrahim wrote. “The documentation is seeds. The forest it might become is beyond our sight. But we plant anyway, because seeds are what we can do.”
The metaphor captured the documentation constructor’s essential character: construction for a future that might never arrive, built on faith that construction was better than silence.
The archive grew. The satellites watched. The testimonies accumulated.
Building a record. Building for tomorrow. Building despite uncertainty about whether tomorrow would ever want what today was building.
Definition: The satellite constructor builds observation from orbit. Its primitives are electromagnetic signatures—light, heat, radar returns—captured by instruments beyond human reach. The satellite constructor sees what ground observers cannot: patterns across space and time invisible from any single location. It is the panopticon repurposed for witness.
The satellites did not choose to document genocide.
They were built for other purposes: weather monitoring, agricultural assessment, urban mapping, military surveillance. They orbited and observed according to programming, indifferent to the moral weight of what their sensors captured.
But their indifference was operationally useful. The satellites couldn’t be threatened, couldn’t be bribed, couldn’t be intimidated. They observed with mechanical reliability, recording whatever fell within their field of view.
The Yale Humanitarian Research Lab had learned to extract testimony from this indifferent observation.
“The satellite doesn’t care about genocide,” a Yale researcher explained. “It just records electromagnetic radiation reflected from Earth’s surface. But we care. And we can interpret what it records. The satellite is our eye; we’re the intelligence that understands what the eye sees.”
The primitives of satellite observation were physical signatures.
Optical imagery: visible light reflected from surfaces, showing colors and shapes. Burned buildings appeared as dark rectangles. Fresh graves appeared as disturbed soil. Blood on concrete appeared as dark staining.
Thermal imagery: heat signatures that revealed active fires, warm bodies, functioning infrastructure. A village being burned showed heat blooms; an evacuated village showed no thermal activity.
Radar imagery: microwave reflections that penetrated clouds and revealed terrain changes. Newly dug trenches showed specific radar signatures regardless of weather.
Multispectral imagery: measurements across multiple wavelengths that revealed things invisible to human eyes. Disturbed soil had different spectral signatures than undisturbed; vegetation stress patterns indicated population movements.
Each primitive captured different information. Combined, they created comprehensive pictures that no single source could provide.
The constructor mechanism was analytical.
Raw satellite imagery was data, not understanding. The construction occurred when analysts interpreted the data—identifying patterns, correlating observations, drawing conclusions.
The Yale team had developed methodologies specifically for humanitarian documentation. They knew what genocide looked like from space: the temporal patterns of village destruction, the spatial distribution of mass graves, the signatures of population displacement.
“We’ve learned to read the Earth’s surface like a text,” the research director said. “Every crater tells a story. Every heat signature has meaning. We’re forensic interpreters of planetary-scale evidence.”
The interpretation required ground truth—comparison with testimonies, coordination with journalists and researchers who could verify what the satellites saw. The satellite constructor worked best when integrated with ground-level documentation.
The constructed output was evidence that couldn’t be destroyed.
Once captured, satellite imagery existed in archives beyond RSF reach. The images were copied to servers in multiple countries. They were preserved in formats that would remain accessible for decades.
The RSF could destroy ground-level evidence—burn documents, bury bodies, level buildings. They could not destroy images already captured from orbit.
“We have the receipts,” a researcher said. “They can bulldoze the mass graves, plant grass over them, pretend they never existed. But we have images from before, during, and after. The evidence is permanent.”
The permanence was a form of construction: building an indelible record that would persist regardless of what happened on the ground.
The satellite constructor revealed patterns invisible from below.
From ground level, each massacre was a local event—this village, these people, this horror. From orbit, the massacres revealed systematic patterns: the geographic progression of violence, the timing correlations, the methodological consistency.
The Yale analysis of El Fasher showed what individual witnesses couldn’t see: the deliberate encirclement strategy, the systematic destruction of specific neighborhoods, the execution sites positioned for efficient processing.
“Ground witnesses know what happened to them,” a researcher observed. “Satellite witnesses know what happened everywhere. The synthesis reveals the system.”
The systemic view transformed understanding. The massacres were not random; they were coordinated. The destruction was not chaotic; it followed patterns. The evidence of system was evidence of intent—crucial for legal prosecution of genocide.
The satellite constructor operated continuously.
Commercial satellites passed over Sudan multiple times daily. Archive imagery extended back years—baseline conditions before the war, progression of changes through the war, current status of affected areas.
The continuous observation created a time-lapse of genocide. The transformation could be watched frame by frame: villages intact in March 2023, fighting visible in April, burned structures in May, abandoned ruins thereafter.
“It’s like watching a slow-motion disaster,” an analyst said. “We can’t stop it. We can only document it. But the documentation is comprehensive. We’re recording every stage of the destruction.”
The comprehensiveness was itself valuable. The record would show not just that destruction occurred but how it occurred—the sequence, the pace, the method. The temporal dimension added to the evidentiary construction.
The satellite constructor had limitations.
Resolution constrained what could be seen. The commercial satellites used by Yale could identify buildings but not individuals. They could detect mass graves but not count bodies precisely. They could observe destruction but not motivation.
“We can see that something happened,” a researcher explained. “We can’t always see exactly what. A dark stain might be blood—or might be oil. A crowd might be perpetrators—or might be refugees. We need ground truth to interpret ambiguous imagery.”
The limitation required integration with other documentation sources. The satellite evidence was strongest when combined with testimony, medical records, and other ground-level data.
Weather constrained observation. Cloud cover blocked optical sensors. Sandstorms degraded radar. The satellites saw what conditions permitted, not what documentation required.
Timing constrained capture. A massacre completed between satellite passes might leave no direct imagery. The evidence of aftermath differed from the evidence of the event itself.
Leila used satellite data in her coordination platform.
The diaspora system integrated Yale’s analysis with other information streams. The imagery showed where displacement was occurring, where infrastructure was destroyed, where populations were concentrating.
“I can see things they can’t see on the ground,” Leila said. “A camp might feel enormous to someone inside it. From above, I can see that another camp nearby is larger, might be more dangerous, might need more resources. The aerial view enables better allocation.”
The integration was imperfect—imagery analysis took time, and conditions changed faster than analysis could capture. But the satellite input improved coordination beyond what ground-level information alone could achieve.
The satellite constructor would continue operating indefinitely.
The satellites would orbit as long as they functioned. New satellites would replace old ones. The observation capacity was effectively permanent.
When the Sudan war ended, the satellites would still be watching. They would document whatever followed: reconstruction, continued conflict, population return, whatever the future held.
“The archive grows forever,” a researcher said. “We’ll be able to compare: what did this area look like in 2020, 2023, 2025, 2030, 2040. The longitudinal record will reveal things we can’t even anticipate yet.”
The permanent observation was both tool and artifact. It served immediate documentation needs and created resources for future analysis. The satellite constructor built for all time.
The satellite constructor was the most alien of the documentation systems.
It was non-human—mechanical sensors operated by software, maintained by engineers who never saw the events they captured. It was distant—orbiting hundreds of kilometers above the suffering it recorded. It was indifferent—observing genocide with the same electromagnetic curiosity it brought to measuring crop health.
Yet it was effective. The inhuman distance that might seem to diminish the horror instead preserved evidence that human witnesses could not. The mechanical indifference that might seem inappropriate was precisely what made the documentation reliable.
“We’ve outsourced part of witnessing to machines,” a researcher reflected. “That’s strange. But the machines are better witnesses in some ways. They don’t forget. They don’t embellish. They don’t die. They just record.”
The satellite constructor was construction at maximum abstraction: building evidence from electromagnetic signals captured by instruments that knew nothing of what they observed. The meaninglessness of the process produced meanings that human observers might never have captured.
The construction continued. The satellites continued circling. The archive continued growing.
Witnessing from orbit. Building from light.
Definition: The future is not given; it is constructed. The present contains multiple possible futures, and which future emerges depends on which construction processes dominate. The RSF constructs one future: ethnic consolidation, permanent displacement, impunity. The resistance constructs another: return, justice, coexistence. Both futures are being built simultaneously. The outcome depends on which construction succeeds.
In December 2025, two months after El Fasher fell, Leila stood before a wall of screens in a London conference room.
The screens showed maps, data streams, network diagrams. Each visualization represented a different aspect of the crisis: population displacement, food distribution, medical needs, documentation status. Together, they formed a picture of a catastrophe in progress.
“We’re tracking approximately fourteen million displaced persons,” she said to the gathered diaspora leaders. “Of those, roughly eight million are internally displaced within Sudan. Six million are in neighboring countries—Chad, Egypt, South Sudan, Ethiopia. The camps are overwhelmed. The food supply is inadequate. The medical infrastructure is collapsing.”
The numbers were abstractions that contained multitudes. Each digit represented thousands of lives—hungry, sick, traumatized, waiting.
“What’s the trajectory?” someone asked. “Where does this end?”
Leila paused. The question had multiple answers depending on which construction process you analyzed.
“The RSF is constructing territorial consolidation. They now control most of Darfur, the Kordofan states, and significant parts of Khartoum. Their goal appears to be permanent demographic change—replacing the populations they’ve displaced with loyal populations. If they succeed, the displaced cannot return. The future they’re building has no place for fourteen million people.”
“And the alternative?”
“The alternative requires that their construction fails. That the displaced survive long enough to return. That justice creates accountability. That the economic and political pressures eventually force settlement. That future is also being built—by the ERRs, by the documentation efforts, by the international advocacy. But it’s being built more slowly and with fewer resources.”
Ibrahim, in Adré camp, was constructing the future through memory.
His documentation center had grown from a single notebook to an organized archive. Testimonies from hundreds of survivors. Maps of massacre sites. Lists of the dead, verified where possible through multiple witnesses. Photographs smuggled out on phones.
The archive was physically vulnerable—paper and hard drives could be destroyed, stolen, lost. But he had created redundancy. Copies in multiple locations. Digital backups distributed through secure networks. The archive existed in fragments across three continents.
“Why do you keep doing this?” a young volunteer asked him one evening. “The war isn’t over. They could still win. Everything you’ve recorded might never matter.”
Ibrahim considered the question. The volunteer was right that outcomes were uncertain. The documentation might never reach a courtroom. The names might never be officially acknowledged. The archive might become nothing more than a repository of grief.
“I’m building a foundation,” he said finally. “I don’t know what will be built on it. Maybe nothing. Maybe everything. But without the foundation, nothing can be built. The future needs raw material—evidence, memory, truth. I’m producing raw material.”
The construction metaphor was apt. A building required materials before it required architects. The documentation was material: bricks of testimony, beams of evidence, mortar of verified truth. What structure might eventually rise was unknown, but the materials would enable possibilities that their absence would foreclose.
Maryam had left El Fasher after the fall.
The escape had been harrowing—days of hiding, a dangerous journey through RSF-controlled territory, eventually reaching Chad with a group of other survivors. She had lost everything except her phone and the photographs it contained.
In the Adré camp, she found work almost immediately. Medical personnel were desperately needed. Her experience during the siege—working with minimal supplies, improvising treatments, handling mass trauma—made her more qualified than most.
But she was constructing more than medical care. She was constructing testimony.
Each patient she treated became a source. “Tell me what happened to you,” she would say, and they would tell her. The injuries spoke of particular weapons, particular methods. The psychological symptoms spoke of particular experiences. She was reading bodies like texts, extracting the story of violence from its physical inscriptions.
“The evidence is walking into my clinic,” she told Ibrahim when they met through the camp’s informal networks. “Every patient is a witness. Every wound is documentation.”
She began systematic recording. Photographs of injuries with patient consent. Notes on mechanisms of harm. Cross-referencing with Ibrahim’s archive to identify patterns—which weapons were used in which locations, which methods were characteristic of which units.
The medical construction and the documentation construction merged. The future being built required both: survival and memory, healing and evidence.
The RSF was also constructing.
Their construction was territorial and demographic. In the areas they controlled, they were resettling populations—moving Arab communities into lands previously occupied by Masalit, Fur, Zaghawa. The new settlements were facts on the ground, physical structures that would be difficult to reverse.
International observers documented the construction with alarm. Satellite imagery showed new buildings in areas recently depopulated. Agricultural patterns shifted as new farmers took over land from those who had fled. The genocide was not only about killing; it was about replacement.
“They’re building a new Darfur,” a UN official said in a leaked report. “They’re erasing the old population and installing a new one. If this construction completes, the displacement becomes permanent. The refugees cannot return to communities that no longer exist, to land that is now occupied by others.”
The RSF commanders understood this strategic logic. Killing was destructive; settlement was constructive. Both served the same goal, but settlement had advantages: it created stakeholders in the new order, populations that would resist any reversal.
“We’re not building for today,” an RSF political officer said in an intercepted communication. “We’re building for fifty years from now. When the children of our settlers are the ones living on this land, whose land is it? History favors those who stay.”
Yusuf had reached Chad and connected with other SAF survivors.
The remnants of the El Fasher garrison—a few hundred men who had escaped the massacre—gathered in camps along the border. They were nominally still soldiers, but without chain of command, without supplies, without clarity about their future role.
Yusuf took unofficial command. His rank, his experience, his survival of the 920-day siege gave him authority that no one challenged.
“We have two options,” he told them. “We can disperse—become civilians, seek asylum, try to build new lives in exile. Or we can organize. Maintain military capacity. Prepare for whatever comes next.”
“What comes next, sir?” someone asked.
“I don’t know. But I know that if we disperse now, we can never reconstitute. If we maintain organization, we preserve options. We can’t control the future, but we can prepare for multiple futures.”
The choice was about construction. Dispersal would construct individual survival at the cost of collective capacity. Organization would construct collective capacity at the cost of individual opportunity. Both were valid constructions, but they led to different futures.
Most stayed. The organizational capacity was maintained. Whether it would ever be used, for what purpose, remained undetermined.
The international community was constructing too, though its construction was largely rhetorical.
The UN Security Council passed resolutions condemning the violence. The US declared genocide. The ICC issued arrest warrants. These were constructions in legal and political space—frameworks that defined what was happening and what responses were legitimate.
But rhetorical construction without material backing produced hollow structures. The resolutions were not enforced. The warrants were not executed. The gold continued flowing through Dubai. The weapons continued reaching the RSF.
“They’re building a paper future,” Leila said bitterly. “A future of statements and declarations. A future that exists in documents but not on the ground.”
Yet even paper futures had potential weight. The legal frameworks being constructed would persist. The precedents being established would shape future cases. If the political conditions ever changed, the rhetorical construction would provide tools for material construction.
“The words matter,” Ibrahim argued when Leila expressed frustration. “Not now, maybe. Not for a long time. But someday, someone will need these words. The legal construction is slow, but it’s real.”
The debate between rhetorical and material construction was itself a form of construction—building the ideological framework through which the crisis would be understood, through which futures would be imagined.
Samira was constructing organizational capacity.
The ERR networks had expanded throughout the crisis, but they remained informal—decentralized, adaptive, resistant to hierarchy. This informality was strength in crisis: it enabled rapid response, local adaptation, resilience against decapitation.
But informality was weakness in construction. Building a future required institutions: stable organizations that could maintain effort over decades, transfer knowledge across generations, accumulate resources and deploy them strategically.
“We need to become an organization,” Samira argued at a coordinator meeting. “Not bureaucratic. Not hierarchical. But structured. We need formal systems for training, for funding, for continuity.”
“That’s what destroyed other movements,” someone countered. “The moment you institutionalize, you lose flexibility. You lose the spirit that made us effective.”
“We’re not institutionalizing to become effective. We’re institutionalizing to become permanent. The crisis won’t end next year. It might not end for a decade. The structures that exist in ten years will shape what’s possible. If we’re not one of those structures, we’re not shaping the future.”
The construction of organizational permanence proceeded slowly, contentiously. Each step toward formalization was debated, negotiated, sometimes rejected. The outcome was uncertain: would the ERRs become an institution that built futures, or would they dissipate when the immediate crisis faded?
The children were construction material of a different kind.
In the camps, a generation was being formed. Children who remembered the genocide. Children who had lost parents, siblings, villages. Children who would carry trauma but also determination into whatever future emerged.
Ibrahim taught classes when he could. Simple lessons: reading, arithmetic, history. The history lessons were the most important—ensuring that the children knew what had happened, understood their history, could articulate the story of their displacement.
“They’ll be the ones who build,” he told Khadija. “Not us. We’re too old, too damaged, too tired. But these children—they’ll have energy we don’t have. They’ll have decades we don’t have. They’ll construct whatever comes next.”
“Or they’ll be broken,” Khadija said. “Trauma doesn’t always produce builders. Sometimes it produces ruins.”
Both possibilities were real. The children were being shaped by experiences that could lead to resilience or to damage, to construction or to destruction. The outcome depended on what happened in the camps—what education they received, what support, what hope.
The battle for the future was partly a battle for children: who shaped them, what they believed, what they became capable of building.
Bashir was also in a camp, though a different kind.
The RSF maintained facilities for fighters rotating out of combat. Rest, resupply, debriefing. Bashir had been sent to one such facility after the El Fasher operation—a reward for his service, a chance to recover before the next deployment.
He did not rest. He thought.
Twenty years of operations. Thousands dead by his hand or under his command. The career trajectory from teenage killer to professional executioner. What had he constructed?
Terror, he supposed. He had constructed fear in communities across Darfur. Fear that enabled displacement, that prevented resistance, that cleared territory for the settlements now rising.
But terror was construction with negative space. He had not built; he had cleared. Others would build on the cleared ground. His role was demolition, not architecture.
Was that enough? He had never asked before. The operations were operations; they had purposes defined by commanders, implemented by soldiers. The purposes were not his concern.
But watching the new settlements rise on land he had cleared, he wondered. What future was he constructing? Whose future? And would that future have a place for him, or would he eventually become debris cleared by someone else?
The thoughts were uncomfortable. He pushed them away. But they returned, persistent, troubling. The construction continued, but his role in it had become unclear.
The multiple constructions proceeded simultaneously.
The RSF built settlements. The ERRs built capacity. The documentation efforts built archives. The survivors built memory. The international community built frameworks.
Each construction had its own logic, its own resources, its own trajectory. They competed and complemented and contradicted each other. The future being built was not singular but multiple—a superposition of possibilities that would collapse into actuality only when time determined which constructions succeeded.
“We’re all builders,” Leila reflected during a rare moment of quiet. “We’re all constructing futures. The difference is what we’re building and who benefits. The RSF builds exclusion. We build inclusion. They build forgetting. We build remembering. The materials are the same—human effort, resources, time. The structures are opposite.”
The construction metaphor had limits. Buildings stood or fell; futures were more complex. A future could partially succeed—some elements realized, others frustrated. The RSF might consolidate territory without achieving complete impunity. The documentation might preserve memory without achieving justice. The ERRs might maintain capacity without achieving return.
The futures being constructed were not binary outcomes but spectra. The question was not which future would win but what combination of elements would eventually stabilize. The construction continued, brick by brick, act by act, choice by choice.
The future remained under construction.
Definition: A manual is a codified construction procedure. It transforms tacit knowledge into transmissible form, enabling replication without direct transmission from practitioner to practitioner. The manual is the constructor’s construction—a mechanism for building more builders. When the ERRs write their manual, they are building the capacity for others to build.
The manual began as a collection of WhatsApp messages.
Samira had been receiving questions from volunteers for months: How do you set up a food distribution point? What’s the protocol for medical evacuation? How do you coordinate with other ERR cells? The questions came from across Sudan, from people who had decided to help but didn’t know how.
Her answers accumulated. She saved them, organized them, noticed patterns. The questions clustered around specific operational challenges. The answers she gave were becoming standardized.
“We’re reinventing the wheel every time someone new joins,” she told Aisha during one of their coordination calls. “I’m answering the same questions over and over. There should be a document.”
“Then write one.”
The suggestion was simple. The execution was not.
The first draft was eighteen pages.
Samira wrote it in two nights, working from her archive of messages. She organized the content around operational phases: how to establish an ERR node, how to assess community needs, how to coordinate resources, how to document activities, how to protect volunteers.
Each section contained procedures refined through practice. The food distribution protocol had been tested during hundreds of distributions, adjusted after each failure, optimized through iteration. The medical coordination procedures came from Maryam and other healthcare workers who had developed them under siege conditions.
“This is our institutional memory,” Samira wrote in the introduction. “This is what we’ve learned. This is how to do what we do.”
She circulated the draft to the experienced coordinators. The feedback was immediate and extensive.
“Section three needs more on security protocols.” “The medical section doesn’t cover mental health.” “What about dealing with armed groups? We encounter them constantly.” “This assumes urban settings. Rural operations are different.”
The eighteen pages grew. The manual was being constructed collaboratively, each contributor adding their experience to the collective knowledge.
The second draft was sixty pages.
The additional content came from across the network. Volunteers in Darfur contributed sections on navigating tribal dynamics. Volunteers in Khartoum added urban-specific protocols. Medical professionals expanded the health sections. Lawyers contributed on documentation for future legal proceedings.
The manual was becoming comprehensive, but it was also becoming unwieldy. Sixty pages was too much for a new volunteer to absorb quickly. The information needed structure.
Leila, with her background in information architecture, helped reorganize. She created a modular structure: core protocols that everyone needed to know, specialized sections that applied to specific roles, appendices with detailed guidance for unusual situations.
“Think of it as a reference, not a textbook,” she advised. “No one reads it cover to cover. They read what they need when they need it. The structure should support quick lookup, not sequential reading.”
The reorganization added another twenty pages of tables of contents, indexes, cross-references. But it made the manual usable.
The manual contained explicit knowledge, but also implicit philosophy.
The protocols were not neutral. They embodied decisions about what the ERRs valued, what they prioritized, what trade-offs they accepted. These decisions were themselves a form of construction—building not just procedures but principles.
“We help everyone regardless of ethnicity, politics, or affiliation.” The statement appeared in the introduction, but it shaped everything that followed. Distribution protocols did not ask about identity. Assessment procedures focused on need, not category. The universalism was operationalized throughout.
“We document everything.” This principle shaped the data collection sections—not just what to record but why. The documentation served immediate operational needs but also future justice. The manual taught volunteers to think on two timescales: the present crisis and the future reckoning.
“We protect our volunteers.” Security protocols were extensive, reflecting the reality that ERR volunteers were targets. The manual taught risk assessment, operational security, emergency procedures. The protection of helpers was itself a form of construction—building capacity for long-term operation rather than burning out quickly.
“We adapt.” The manual was explicitly not final. It contained version numbers, revision dates, procedures for submitting updates. The construction was ongoing, the manual itself under continuous development.
Translation expanded the manual’s reach.
The original was in Arabic, but Sudan’s linguistic diversity demanded more. Volunteers translated sections into Fur, Masalit, Zaghawa, Beja, Nubian. Each translation required adaptation—not just linguistic but cultural. Concepts that made sense in Arabic needed rephrasing to resonate in other contexts.
The translation process revealed assumptions. Phrases that seemed universal turned out to be culturally specific. Organizational structures that felt natural in some communities were alien in others. The translators were not just converting words but reconstructing the manual for different audiences.
“This is harder than I expected,” a Fur translator reported. “The concepts are fine, but the framing is very Khartoum. Rural communities organize differently. I’m not just translating; I’m rewriting.”
The rewriting was permitted, even encouraged. The manual’s construction was distributed, each community building its own version while maintaining fidelity to core principles.
The digital version enabled distribution at scale.
PDF files circulated through WhatsApp, Telegram, email. The manual reached volunteers who had never met anyone from the original ERR networks, who were constructing responses in isolation based solely on the document.
Reports came back from these independent implementers. Success stories: a village that established food distribution using the protocols, maintaining it for months. Failure stories: a group that tried to apply urban procedures in rural settings, encountered problems the manual didn’t anticipate.
The feedback loop continued. Each failure was analyzed, and if the failure was general—if it might happen to others—the manual was updated to prevent repetition. The construction was iterative, improving through use.
“We’re debugging our own operations,” Aisha observed. “Every error in the field is a bug in the manual. We fix the manual so future volunteers don’t hit the same bugs.”
Ibrahim contributed a section on documentation for justice.
His experience in collecting testimonies had taught him what evidence was needed, how to collect it, how to preserve it. This knowledge was encoded in the manual’s documentation protocols.
“When you record a testimony,” the section instructed, “note the time, location, and your own presence. Describe the witness’s condition. Ask open-ended questions: what happened, when, where, who was involved. Avoid leading questions. Record exactly what they say, even if it seems confused or contradictory. The confusion might be significant later.”
The instructions were forensic in their precision. They anticipated the needs of prosecutors, investigators, historians. They taught volunteers to think like archivists, preserving information for uses that might not emerge for years.
“This section is for the future,” Ibrahim wrote in a sidebar. “You may never see the results of this documentation. But someone will. The care you take now determines what is possible later.”
The RSF had no equivalent manual.
Their operations were transmitted through command hierarchy, through training by experienced fighters, through the accumulated tradition of the Janjaweed. The knowledge was tacit, embedded in people rather than documents.
This difference had implications. The RSF’s methods could not scale beyond the reach of experienced practitioners. Each new recruit needed individual training. The organizational knowledge was fragile, concentrated in individuals who could be killed or captured.
The ERR’s manual enabled geometric expansion. One volunteer with a manual could train ten others. Those ten could each train ten more. The knowledge scaled without requiring proportional expansion of experienced practitioners.
“They have guns,” Samira said. “We have documents. It doesn’t seem like a fair trade. But guns require supply chains; documents require only copying. In the long run, the copyable thing wins.”
The analysis was optimistic, perhaps overoptimistic. But it captured a real asymmetry. The manual was a replication mechanism—a way of building builders—that the RSF’s approach could not match.
Maryam contributed the medical protocols.
Her experience in siege medicine—improvising, rationing, triaging—became codified guidance. The protocols assumed scarcity: limited supplies, limited personnel, limited time. They taught prioritization under impossible conditions.
“You cannot save everyone. This is the first thing you must accept. If you try to save everyone, you will save fewer than if you accept limits and prioritize.”
The brutal truth was stated plainly. The manual did not pretend that humanitarian work was clean or heroic. It prepared volunteers for the moral weight of choices they would make.
“Document the ones you cannot save. Their names, their conditions, the reason you could not help. This documentation is itself a form of help—it preserves them as individuals rather than statistics. It is the minimum dignity you can offer.”
The medical section was the hardest to write and the hardest to read. It constructed capacity for work that would damage those who did it. It included sections on self-care, on recognizing burnout, on when to stop.
The manual was never finished.
Each version was superseded by the next. Updates flowed continuously as the crisis evolved and new lessons accumulated. The construction was perpetual, the document a living artifact.
By December 2025, the manual was in its seventeenth major version. It had been distributed to an estimated fifteen thousand volunteers across six countries. It had been translated into twelve languages. It had spawned derivative documents—specialized manuals for specific contexts, condensed versions for quick training, expanded versions for reference.
The construction it enabled was invisible from the outside. No one saw the manual when they saw ERR operations. They saw the food distributions, the medical clinics, the documentation efforts. But beneath the visible operations was the structure the manual had built—the shared understanding, the standardized procedures, the institutional knowledge that made coordination possible.
“We built a machine for building machines,” Leila said, reviewing the distribution statistics. “The manual makes ERR nodes. The nodes make ERR operations. The operations make survival. We’re several layers up from the direct impact, but we’re still part of the construction.”
The manual was the constructor’s construction: a mechanism for generating mechanisms for generating help. Its abstraction was its power. The construction continued, version by version, update by update, volunteer by volunteer.
Part VI was complete. The universal construction had been examined from every angle.
Now the system would turn to examine itself.
Definition: Recursion occurs when a process takes itself as input. The witness witnesses their own witnessing. The documenter documents the documentation. The system’s self-reference creates strange loops—feedback that transforms the system by making it aware of itself. Part VII examines this self-examination: the automaton turning to analyze itself.
Ibrahim found himself reading his own notebooks.
It was March 2026. The Adré camp had stabilized into something resembling permanence—semi-permanent structures replacing tents, an administrative order emerging from chaos. His documentation center had grown into a formal archive, staffed by volunteers, visited by journalists and researchers.
He sat in the quiet of early morning, before the day’s activities began, and read what he had written.
June 16, 2023. Testimony of Amina Hassan, age 34. Witnessed the killing of her husband and two sons in the market of El Geneina. Describes uniformed men with RSF insignia. Estimated time: approximately 10:30 AM.
The words were precise, clinical. He had recorded them with the detachment necessary for documentation—focusing on facts, suppressing emotion, capturing what could be verified.
But reading them now, he felt the weight he had suppressed. Amina Hassan. He remembered her face, her voice breaking as she spoke, the way her hands shook. The notebook contained data; his memory contained a human being.
“I was becoming a recording device,” he wrote in his personal journal—the one not meant for the archive, the one where he examined himself. “I trained myself to capture information without experiencing it. This was necessary. But what did it cost?”
The witness witnesses their own transformation.
Ibrahim had been a teacher. He had taught children to love Arabic literature, to appreciate the rhythms of the language, to find joy in learning. The classroom had been his natural environment, the place where his skills and temperament aligned.
Now he was something else. A documentarian. An archivist. A professional recorder of atrocity. The skills he had developed were useful—organization, precision, patience—but the context had transformed them into something he had never intended.
“Who am I now?” he wrote. “Not the teacher. Not anymore. The teacher is still inside, somewhere, but buried under layers of testimony. I am the person who knows what happened in El Geneina. I am the carrier of this knowledge. Is that an identity?”
Khadija found him writing one evening and read over his shoulder.
“You’re spiraling,” she said. “I’ve seen this before. In myself.”
“You?”
“After 2003. When the first genocide was over and I had to decide what to do with what I’d seen. I spent months examining myself—what I had become, what I had lost, whether the person who survived was still the person I had been.”
“And?”
“She wasn’t. The old Khadija was gone. A new Khadija had emerged. Scarred, different, but still Khadija. Still capable of love, of hope, of building a life. The continuity wasn’t perfect, but it was sufficient.”
Ibrahim considered this. The recursion—examining his own examination—had a purpose. It was not merely self-indulgence. It was the work of integration, of constructing a self that could contain both the teacher and the documentarian, both the joy of literature and the weight of testimony.
The documentation center received its own documentation.
Journalists came and photographed the archive, the volunteers, Ibrahim at work. They wrote stories about “the man who records genocide,” framing his work as heroic, singular, remarkable.
He read these stories with discomfort.
“They’re missing something,” he told Khadija. “They see one man with notebooks. They don’t see the network—the hundreds of people who provide testimonies, the volunteers who transcribe and organize, the diaspora connections that preserve and distribute. They make it about me because individuals are easier to narrate than systems.”
The meta-documentation—documentation of documentation—introduced distortions. The narrative simplified what was complex, individualized what was collective, heroized what was simply necessary.
“Let them tell it wrong,” Khadija suggested. “What matters is that they tell it at all. The detailed truth lives in the archive. The simplified version brings attention, resources, support. Both are needed.”
Ibrahim accepted the logic but remained uncomfortable. The witness being witnessed, the documenter being documented—the recursion created versions of himself that he didn’t recognize. The Ibrahim of the newspaper stories was a character, not a person.
He began documenting the documentation process itself.
It started as a methodological project: recording how testimonies were collected, what questions were asked, what biases might shape the archive. This meta-documentation would help future researchers understand the context of what they were examining.
But it evolved into something more personal. He documented his own emotional state during interviews, his reactions to particular testimonies, the cumulative effect of months of exposure to trauma. He documented the volunteers—their motivations, their struggles, their transformations.
“We are artifacts too,” he wrote. “The genocide shaped us as surely as it shaped the survivors we interview. The archive should contain not just testimonies of victims but testimonies of witnesses. We are evidence of what this has done.”
The recursion deepened. Documentation of documentation of genocide. Layers of self-reference that could extend infinitely.
Where to stop?
The practical limit was capacity.
Ibrahim had time for only so much reflection. The ongoing crisis demanded attention: new testimonies arriving daily, new volunteers needing training, new challenges in preservation and distribution. The recursion had to be bounded by necessity.
He established a routine. Early mornings for personal reflection, for reading old notebooks, for examining his own role. Working hours for the operational tasks—interviewing, organizing, coordinating. Evenings for processing, for conversation with Khadija and colleagues, for the informal exchange of perspectives.
The routine created structure within which recursion could occur without overwhelming. The self-examination was contained, a component of work rather than a replacement for it.
“I think about what I’m doing,” he explained to a visiting researcher. “I think about why I’m doing it and what it means. But I don’t let the thinking stop the doing. The thinking shapes the doing; the doing provides material for thinking. They have to coexist.”
The researcher nodded and took notes. She was documenting his documentation, adding another layer to the recursion.
Maryam underwent her own recursive process.
In the camp clinic, between patients, she found moments to examine what she had become. The nurse who had worked in Khartoum’s hospitals—efficient, skilled, focused on individual patients—was now something else. A humanitarian worker. A documentarian of physical trauma. A witness whose testimony was inscribed in the injuries she treated.
“The bodies are texts,” she wrote in her own journal. “I read them, I interpret them, I record what they say. But I’m also a text—shaped by reading so many bodies, marked by the cumulative exposure. What would someone read in me?”
The question was uncomfortable. She had become harder, more detached, more capable of viewing suffering with clinical distance. These were adaptive changes, necessary for function. But they were also losses—the erosion of something softer, more connected, more fully human.
“I’m not sure I can go back,” she confessed to a colleague. “To normal medicine, I mean. Routine care, ordinary patients, conditions that don’t involve war crimes. It would feel trivial. I’ve been shaped by this, and the shaping is permanent.”
The colleague understood. They had all been shaped. The recursion—examining their own transformation—was universal among the long-term workers.
Leila, in London, experienced a different kind of recursion.
She analyzed data about the crisis, but she also analyzed her own analysis. What patterns was she seeing? What patterns was she missing? How did her distance—geographic, emotional, experiential—shape what she could perceive?
“I’m looking at Sudan through screens,” she wrote in a reflection for her own records. “Maps and numbers and network diagrams. The crisis is abstracted into information, and I process the information with tools built for fraud detection. Is this insight or blindness?”
The question had no clear answer. Her distance provided perspective that those on the ground lacked—she could see patterns across regions, correlations across time, connections that local observers might miss. But her distance also created blind spots—the human reality that no data could capture, the suffering that statistics could only approximate.
The recursion for her was methodological: examining the methods, questioning the assumptions, acknowledging the limits. The analyst analyzing analysis.
Even Bashir, in his way, experienced recursion.
The thoughts that had troubled him after El Fasher—questions about what he was building, whose future, his own role—refused to fade. In the rest facility, surrounded by other fighters, he found himself watching himself.
“I see myself as if from outside,” he mentioned to a fellow veteran. “I see Bashir the soldier, Bashir the executioner, Bashir the man who follows orders. And I wonder who is watching.”
The fellow veteran looked at him strangely. “That sounds like philosophy. We don’t do philosophy. We do operations.”
But the recursion continued. Bashir could not stop the self-observation, could not return to the unreflective execution that had characterized his earlier career. Something had shifted—perhaps the scale of El Fasher, perhaps accumulated weight of twenty years, perhaps simply age bringing perspective.
“The machine examines itself,” he muttered, not knowing that he was anticipating the next chapter’s theme. “And what does it see?”
What it saw was troubling. The recursion, for him, was not clarifying but destabilizing. The self-examination threatened the unreflective efficiency that had enabled his function.
Ibrahim’s ultimate recursion was about meaning.
Why document? The question returned repeatedly in his reflections. The immediate answers were clear: for justice, for memory, for history. But beneath these instrumental purposes lay deeper questions.
“I document because the world should know,” he wrote. “But the world already knows. The information is available. The genocide is acknowledged. And yet nothing changes. So what is documentation for?”
The question was not rhetorical. It required serious engagement.
“Perhaps documentation is not about changing the world. Perhaps it is about asserting reality—insisting that what happened, happened, regardless of whether the world responds. The assertion is itself valuable. It is a refusal to let them erase. Even if no court ever convicts anyone, even if no government ever intervenes, the documentation says: this occurred. You cannot pretend otherwise.”
The recursion led to this insight: the documentation was not only for external effects but for internal ones. It preserved the sanity of witnesses by giving them meaningful work. It preserved the dignity of victims by treating their deaths as significant enough to record. It preserved the possibility of future justice by maintaining the evidence that future generations might use.
“I document for the future,” Ibrahim concluded. “Not the future I will live to see, but the future someone will live to see. The long future, the eventual future, the future in which these notebooks might matter. I’m building for that future even if I can’t see it.”
The witness witnessed himself witnessing, and found something unexpected: resilience.
The recursion—the endless self-examination—did not lead to paralysis or despair. It led to understanding. Ibrahim understood himself better than he had before the war. He understood his limits, his capacity, his purpose.
“I know what I am now,” he wrote in the final entry of this period of reflection. “I am a memory keeper. Not a teacher anymore, though I teach when I can. Not a fighter, though I resist in my way. A keeper of memory. This is what the crisis made me, and this is what I can contribute.”
The identity was not chosen freely; it was constructed by circumstance. But the acceptance of it was his own choice. The recursion had produced not confusion but clarity.
The witness witnesses, and in witnessing their own witnessing, finds themselves.
Definition: A machine that examines itself encounters paradoxes. The RSF, as a machine for genocide, develops mechanisms of self-justification—narratives that explain its operations to itself. These narratives are not external propaganda but internal sense-making. The machine must believe in its own purpose to function effectively.
The RSF had a political wing.
It was called the Rapid Support Forces Political Affairs Office, and its function was to articulate what the RSF was doing and why. The office produced communiqués, managed relations with potential allies, constructed narratives for domestic and international consumption.
But the narratives were not merely instrumental. The officers who crafted them believed what they wrote—or believed enough to function. The machine was examining itself and producing self-descriptions that enabled continued operation.
“We are restoring order,” a senior political officer explained in a document captured by SAF forces and eventually shared with researchers. “The old Sudan was chaos—ethnic militias, rebel groups, a government that couldn’t control its territory. We are creating a new Sudan. A Sudan with clear authority, clear boundaries, clear identity.”
The narrative was coherent. It reframed the genocide as nation-building, the massacres as necessary force, the displacement as consolidation. The reframing was not cynical; it was sincere—or sincere enough.
Bashir encountered the political narratives through training materials.
The RSF had begun formal indoctrination for its fighters, recognizing that motivation required more than payment. Soldiers who understood their mission as righteous fought more effectively than soldiers who saw themselves as mercenaries.
“We are the protectors of Sudan,” the materials declared. “The SAF was corrupt, weak, unable to defeat the rebels who threatened our nation. We rose from the people—from the tribes who had always defended this land—to do what the army could not. Our victory is Sudan’s victory. Our enemies are Sudan’s enemies.”
The framing was tribal and national simultaneously. The RSF fighters were positioned as authentic Sudanese, rooted in tradition, defending against external threats and internal traitors. The ethnic cleansing was recast as patriotic action.
Bashir read the materials with mixed feelings. Part of him recognized the propaganda—he had been doing this work for twenty years without needing ideological justification. But part of him was grateful for it. The narrative made the work feel meaningful, purposeful, connected to something larger than personal survival.
“Maybe they’re right,” he found himself thinking. “Maybe this is what Sudan needs. Maybe the killing is building something.”
The machine’s self-examination had produced self-justification, and the self-justification was colonizing his consciousness.
The RSF also examined its operations analytically.
After-action reviews assessed what had worked and what had not. El Fasher’s siege had been tactically successful but costly in time and resources. The El Geneina operations had achieved rapid ethnic clearing but attracted international attention. Each operation produced lessons.
“We need to control information better,” a review document concluded. “The satellite imagery is damaging. We should conduct operations during cloud cover when possible. We should accelerate body disposal to reduce visible evidence.”
The analysis was coldly operational. The genocide was treated as a project to be optimized, with inputs and outputs and efficiency metrics. The machine examined itself not morally but functionally: how to achieve objectives with minimum friction.
This technical self-examination was perhaps more disturbing than the ideological self-justification. The ideology at least engaged with questions of right and wrong, even if its answers were monstrous. The technical analysis treated moral questions as irrelevant—outside the scope of optimization.
Some RSF members examined the machine and found it wanting.
Not many. The organizational culture punished dissent, and the incentive structures rewarded compliance. But some—particularly younger conscripts like Tariq who had not chosen to join—developed internal critiques.
Tariq kept no written record of his doubts. Writing was dangerous; notes could be found. But he thought, in the privacy of his mind, about what he was part of.
“We are killing people for being who they are,” he formulated, silently, during operations. “Not for what they’ve done. For their ethnicity, their ancestry, their names. This is wrong. This has always been wrong. No narrative can make it right.”
The internal critique did not produce external action. Tariq continued following orders, continued participating in operations, continued being a component of the machine. The self-examination changed nothing visible.
But it changed something invisible. Tariq was no longer a fully integrated component. He was a component that knew itself to be in a machine it did not endorse. This knowledge was itself a form of resistance, even if it produced no immediate effect.
“Someday,” he promised himself, “I will testify. I will tell what I saw and what I did. The machine will be examined from inside, by someone who was inside.”
The international observers examined the RSF from outside.
Think tanks produced analysis. UN experts wrote reports. Journalists investigated supply chains. Each examination added to the understanding of how the machine worked.
“The RSF is best understood as a business enterprise,” one analysis concluded. “Its core business is violence, but violence is deployed strategically to capture resources—gold, land, international payments. The ideological and ethnic dimensions are real but secondary to the economic logic.”
Another analysis disagreed: “The RSF is primarily an ethnic project. The economic resources support the project, but the project’s goal is demographic transformation—creating an Arab-dominated Darfur, eventually an Arab-dominated Sudan. The business analysis misses the ideological core.”
The examinations contradicted each other, each capturing some aspect of the machine while missing others. The machine was complex enough to support multiple interpretations, each partially accurate.
Hemedti, the RSF’s leader, examined his creation with pride.
In speeches and interviews—rare, carefully managed—he presented himself as a builder, a modernizer, a man who had transformed tribal militias into a professional force. The transformation was real, even if the profession was genocide.
“We started with nothing,” he said in one interview, recorded before the war intensified and international access ended. “Horsemen with old rifles. Now we have technicals, drones, supply chains across three countries. We built this from the ground up. No one gave it to us. We made it ourselves.”
The pride was genuine. Whatever moral evaluation one applied to the RSF, it was an organizational achievement. Building a force capable of sustained operations across thousands of kilometers, maintaining discipline and logistics and command, required competence. The machine worked.
Hemedti’s self-examination was the founder’s perspective: satisfaction in creation, investment in continuation, blindness to moral critique that threatened the project’s legitimacy. The machine examined itself through its creator’s eyes and found itself good.
The machine’s self-examination revealed something important about machines.
Machines do not examine themselves spontaneously. They examine themselves when examination is functional—when it improves performance, maintains morale, addresses threats. The RSF’s self-examination served these purposes. It did not serve the purpose of moral critique because moral critique was not functional for the machine’s operation.
This was why external examination mattered. The documentation Ibrahim maintained, the analysis Leila conducted, the reporting journalists produced—these examined the machine from perspectives it could not adopt internally. The external examination held possibilities the internal examination foreclosed.
“They cannot see what they are,” Ibrahim observed. “Or they can see it but cannot accept it. The self-justification is too strong, too necessary for their continued function. Only from outside can the machine be truly seen.”
The observation had implications for intervention. If the machine could not reform itself—if its self-examination produced only optimization and justification—then change had to come from outside. External pressure, external constraints, external consequences.
The machine would not examine itself into morality. It had to be stopped by something other than self-reflection.
The ERR networks also examined themselves, but differently.
Their self-examination was genuinely critical. Coordination meetings included discussions of what had gone wrong, what biases had affected decisions, what could be improved. The culture permitted—encouraged—honest assessment of failures.
“We’re losing people to burnout,” Samira acknowledged in one meeting. “We’re asking too much, not providing enough support. This is a systemic failure, not individual weakness. We need to restructure.”
The self-examination led to changes. Support systems were strengthened. Rotation schedules were adjusted. The machine examined itself and modified itself based on what it found.
This capacity for genuine self-criticism was the difference between a machine that optimizes for its goal and a machine that questions its goal. The RSF optimized genocide. The ERR questioned its own operations, including the fundamental questions of whether it was achieving its purposes and whether its purposes were correct.
The recursive capacity—examining not just performance but purpose—was what made the ERR something more than a machine. It was an organization with agency, capable of changing direction based on reflection.
The two self-examinations—RSF and ERR—produced different futures.
The RSF’s self-examination produced efficiency without conscience. The machine became better at genocide, learning from each operation, optimizing its methods. But it could not become moral; morality was outside its optimization function.
The ERR’s self-examination produced adaptation with conscience. The network became better at aid delivery, but also questioned whether its approach was correct, whether its values were sound, whether its structure served its stated purposes.
The difference was not in intelligence—both systems were capable of learning. The difference was in what they learned about. The RSF learned how to kill more effectively. The ERR learned how to help more effectively and also whether helping in this way was right.
“We are what we examine ourselves into,” Leila wrote in a reflection. “They examine themselves and find efficiency. We examine ourselves and find... complexity. Doubt. But also clarity about what matters. The examination shapes the examiner.”
The machine that examines itself confronts the limits of self-knowledge.
The RSF could not fully see what it was because fully seeing would make operation impossible. The self-justifications were necessary illusions, load-bearing narratives that enabled continued function.
The ERR could see itself more fully because its operation did not require illusion. Helping people was defensible on examination; killing people was not. The difference in activity produced a difference in self-examination capacity.
But both faced limits. The RSF could not see its own evil. The ERR could not see its own blind spots—the assumptions it could not question because they were too foundational, the biases it could not recognize because they were too pervasive.
Total self-knowledge was impossible for any machine, including the human minds that comprised these machines. The recursion was always incomplete, the self-examination always partial.
What mattered was whether the partial examination moved toward truth or away from it. The RSF’s examination moved away—each iteration of self-justification added another layer of distortion. The ERR’s examination moved toward—each iteration of self-critique removed another layer of illusion.
The direction mattered more than the completeness. No machine could fully know itself. But some machines could know themselves better over time, while others knew themselves worse.
The machine examines itself, and the quality of the examination determines the quality of the machine.
Definition: The archive is memory institutionalized. It transforms ephemeral testimony into durable record, individual recollection into collective knowledge. The archive is a recursive structure: it contains documents about events, documents about documents, documents about the archive itself. It is designed to outlast those who create it.
The Sudan Atrocity Documentation Center opened in Nairobi in June 2026.
The location was practical: Kenya was accessible, relatively stable, and had a Sudanese diaspora community that could staff and support the operation. The physical facility was modest—a converted warehouse with climate-controlled storage, computer workstations, and a small reading room.
But the archive it housed was vast.
Ibrahim’s notebooks had been digitized and cross-referenced. Maryam’s medical documentation had been organized by location, date, and type of injury. Leila’s data systems had been consolidated into a queryable database. The Yale HRL satellite imagery had been tagged and integrated. The ERR’s operational records had been preserved.
Together, these sources comprised the most comprehensive documentation of the Sudan genocide yet assembled.
“We’re building memory,” the center’s director said at the opening ceremony. “Memory that doesn’t depend on any individual, that can’t be destroyed by burning a single building or killing a single witness. Distributed, redundant, permanent.”
The archive’s organization was itself a form of interpretation.
Documents could be arranged by date, by location, by type, by subject. Each arrangement told a different story. The archivists had to choose, and their choices shaped what future researchers would find.
Ibrahim participated in the organizational discussions, advocating for victim-centered arrangement.
“The testimonies should be organized by victim, not by perpetrator,” he argued. “Each person who was killed or displaced is a complete story. The perpetrators are secondary—instruments of a system. The victims are primary—the purpose of our documentation.”
Others disagreed. “For prosecution, we need perpetrator-centered organization. Who gave orders? Who executed them? The chain of command is what courts need.”
The debate was not resolved; it was accommodated. The archive implemented multiple access pathways, allowing the same documents to be found through victim searches, perpetrator searches, location searches, date searches. The organization was multidimensional, reflecting the multiple purposes the archive served.
The archive attracted users with different needs.
Families came to find information about the dead. A mother might search for any record mentioning her son’s name, hoping to learn his fate. The archive staff developed protocols for these searchers—gentle guidance, private spaces for emotional responses, connections to support services.
Journalists came to research stories. They needed specific incidents, quotable testimonies, verifiable facts. The staff learned to assess journalistic purposes, distinguishing those who would use the material responsibly from those seeking sensationalism.
Legal investigators came to build cases. The ICC team spent weeks in the archive, extracting evidence for prosecutions. They needed chain of custody documentation, authentication of sources, correlation across multiple witnesses.
Scholars came to study the genocide systematically. They needed aggregate data, statistical patterns, comparative frameworks. Their time horizon was longer; their needs were analytical rather than immediate.
Each user type required different support. The archive was not merely a repository but a service, actively mediating between its contents and those who sought them.
The archive preserved not only content but context.
A testimony was more than words. It was words spoken by a particular person, in a particular condition, at a particular time. The same words from different sources meant different things. The archive tried to preserve this context.
“We record everything we can about the circumstances of testimony collection,” an archivist explained to a visiting researcher. “Who collected it? What questions were asked? What was the witness’s condition? Were there translation issues? These metadata are as important as the testimony itself.”
The contextualization was itself documented, creating layers of meta-information. The archive contained documents about documents about events. The recursion was functional: it enabled users to assess reliability, to understand limitations, to interpret appropriately.
Digital preservation presented challenges.
File formats became obsolete. Storage media degraded. Software that could read old files was discontinued. The archive had to maintain not only data but the ability to access data across technological generations.
“We’re fighting entropy,” the technical director said. “Digital information is not permanent. It requires active maintenance—migration to new formats, refreshing of storage media, updating of access systems. The archive is not static; it’s an ongoing process.”
The process was funded uncertainly. Grants came and went. Staff turned over. Each transition threatened continuity. The archive had to be designed for resilience, for operation despite funding gaps and personnel changes.
They implemented redundancy: multiple copies in multiple locations, multiple formats for each document, multiple access systems. The redundancy created costs but ensured survival. The archive could lose any single component and continue functioning.
The archive faced hostile attention.
The RSF had reason to fear what the archive contained. Evidence that might eventually support prosecutions. Testimonies that contradicted their narratives. Documentation that would shape historical memory.
Cyber attacks targeted the archive’s digital systems. Physical surveillance monitored the Nairobi facility. Threats reached staff members through intermediaries. The archive was not merely preserving history; it was contesting the future, and those who preferred a different future noticed.
Security became a priority. Digital systems were hardened. Physical access was restricted. Staff were trained in operational security. The archive operated partly in shadow, revealing its existence while concealing its defenses.
“We’re a target because we matter,” the director observed. “If the archive were irrelevant, no one would attack it. The attacks are evidence of significance.”
Ibrahim visited the archive in August 2026.
Walking through the facility, seeing his notebooks displayed in climate-controlled cases, he experienced a strange sensation. The documentation he had created was no longer his. It had become institutional, absorbed into a structure that would outlast him.
“This is what I was building toward,” he said to Khadija, who had accompanied him. “I didn’t know it at the time. I was just recording what I saw. But this—this permanence—this was the purpose.”
“Are you happy about it?”
He considered. “Happy is the wrong word. Satisfied, maybe. The memory will persist. Even if I die tomorrow, this will continue. The witness has been witnessed, and the witnessing has been preserved.”
The recursion was complete, in a sense. The archive archived the archiving. The documentation documented the documentation. The layers of self-reference created a structure resilient enough to survive its creators.
The archive’s existence changed what was possible.
Before the archive, the genocide’s documentation was scattered—fragments held by individuals, organizations, governments. Researchers had to assemble their own collections, duplicating effort, missing connections.
After the archive, a comprehensive resource existed. Anyone studying the genocide could start here, building on the accumulated work of thousands. The archive enabled synthesis that fragmentation had prevented.
“We’ve changed the epistemology,” a scholar observed. “The question is no longer ‘is the genocide documented?’ It is. The question is now ‘what do we do with the documentation?’ That’s a different kind of question. A better kind.”
The shift from documentation to interpretation marked a transition. The archive had not completed the work; it had enabled the next phase. The genocide was recorded; now it could be understood.
The archive also preserved the ERR’s records.
The operational documents—coordination protocols, distribution logs, volunteer networks—were as significant as the atrocity documentation. They recorded not only what was done to people but what people did for each other.
“This is a counter-archive,” Samira noted during her visit. “Not just evidence of crime but evidence of response. Future generations will see both: the genocide and the resistance, the killing and the helping. The full picture.”
The counter-archive had its own uses. It provided a model for future humanitarian responses. It documented what worked and what didn’t. It preserved the institutional memory of a movement that might otherwise fade.
The archive was not neutral. By including both genocide documentation and ERR records, it made an editorial statement: both mattered. The violence and the response were equally worthy of preservation.
The archive was designed to outlast its context.
Someday, the Sudan crisis would end—through resolution, exhaustion, transformation. The political configurations that made the archive necessary would change. New priorities would emerge. The archive might seem less urgent.
The designers planned for this. The archive was structured not only for current use but for historical preservation. Future scholars, future courts, future generations seeking to understand what had happened—the archive was built for them as much as for present users.
“We’re writing to the future,” the director said. “We don’t know who will read this. We don’t know what questions they’ll ask. We can only preserve as much as possible, organize it as clearly as possible, and trust that the future will find uses we can’t anticipate.”
The trust was an act of faith. The archive was a message in a bottle, thrown into the ocean of time, hoping someone would receive it.
The recursion extended into the indefinite future. The archive contained documents about the past, organized by present understandings, preserved for future interpretation. The layers of time and meaning accumulated, each generation adding its own layer to the structure.
Memory institutionalized. Witness preserved. The archive as the closest thing to permanence that human effort could achieve.
Definition: The court is a machine for processing evidence into judgment. It takes inputs—testimony, documents, analysis—and produces outputs—verdicts, sentences, precedents. The court is not primarily about the past; it is about establishing what the past means for the present and future. It is recursion formalized: the system judging the system.
The International Criminal Court had been watching Sudan since 2005.
The first Darfur genocide had produced referrals, investigations, arrest warrants. Omar al-Bashir had been indicted. Other commanders had been named. The legal machinery had operated, slowly and incompletely, processing the first genocide while the second unfolded.
When the 2023 war began, the ICC activated its existing investigation mandate. The same apparatus that had pursued 2003 perpetrators now turned to 2023 perpetrators. The continuity was both institutional and personal—some of the same judges, prosecutors, and investigators.
“We’re not starting from zero,” the lead prosecutor said in a statement. “This is a continuation of a case we’ve been building for twenty years. The perpetrators are largely the same people. The methods are largely the same methods. What’s new is the scale and the evidence.”
The evidence was unprecedented in volume. The archive in Nairobi, the satellite imagery from Yale, the diaspora documentation networks—the court had never had such comprehensive material for any case.
The court’s structure shaped what it could do.
ICC jurisdiction was limited. It could prosecute individuals, not states. It could address specific crimes—genocide, crimes against humanity, war crimes—not general injustice. It could only act where national courts were unwilling or unable, and Sudan’s collapse had created that condition.
The limitations were frustrating to those who wanted comprehensive justice.
“The court can prosecute Hemedti,” a legal analyst explained. “Maybe other commanders. But it cannot prosecute the UAE for funding the genocide. It cannot prosecute the gold traders who enable the RSF. It cannot address the systemic conditions that made genocide possible.”
Ibrahim felt this frustration acutely. The archive he had built contained evidence of systems, not just individuals. The gold supply chain. The weapons networks. The international indifference. The court could use his documentation but could not address the full scope of what his documentation revealed.
“It’s better than nothing,” Khadija counseled. “A few prosecutions establish truth even if they don’t achieve justice. The court says officially: this happened. That matters.”
The first arrest warrant was issued in December 2026.
It named Abdel Rahim Dagalo, a senior RSF commander and Hemedti’s brother. The charges: crimes against humanity and war crimes committed in El Geneina. The warrant was public, announced with fanfare, reported worldwide.
The practical effect was negligible. Dagalo was in Sudan, in areas the RSF controlled. No one could arrest him. The warrant was symbolic—a statement of what should happen rather than an action that would happen.
But symbols had power. The warrant marked Dagalo as an accused war criminal. It restricted his travel—any country party to the Rome Statute would be obligated to arrest him. It signaled that impunity was not guaranteed.
“He can never leave Sudan safely,” a legal analyst noted. “He can never travel to London or Paris or any of dozens of capitals. The warrant shrinks his world. It doesn’t put him in prison, but it creates a prison of geography.”
The court required witness testimony.
Investigators came to the camps, to the diaspora communities, seeking those who could testify. The testimony had to meet legal standards—reliability, specificity, chain of evidence. The informal documentation Ibrahim had collected had to be transformed into formal witness statements.
Maryam was contacted as a potential witness. Her medical documentation, her presence in El Fasher during the siege, her observations of the massacre—all relevant to multiple charges.
“They want me to testify,” she told Ibrahim. “To go to The Hague, sit in a courtroom, describe what I saw.”
“Will you?”
“I’m afraid. Not of the testimony itself—I’ve told this story many times. But of being identified. My family is still in Sudan. If the RSF knows I’m testifying...”
The fear was realistic. Witnesses had been threatened, killed, silenced. The court offered protection, but protection had limits. Maryam would have to choose between contributing to justice and protecting her family.
She chose to testify, eventually, after months of deliberation. Her family was moved to safety first, a complex extraction coordinated through diaspora networks. Only then did she travel to The Hague.
The trial, when it came, was years in the future.
International justice moved slowly. Investigation, indictment, arrest (if it ever happened), pre-trial proceedings, trial, appeals—each phase could take years. The 2003 Darfur case had been ongoing for two decades without significant convictions.
“Justice delayed is justice denied,” the activists complained.
“Justice rushed is justice failed,” the lawyers responded. “Due process takes time. The alternative is show trials.”
The debate was irresolvable. Both positions were correct. The speed of atrocity vastly exceeded the speed of justice. By the time any verdict arrived, the genocide would be history—settled, irreversible, the perpetrators old or dead.
But the slowness was also protective. It allowed thorough investigation, careful analysis, proper defense. The court’s legitimacy depended on its process, even if the process frustrated those seeking immediate accountability.
Yusuf was contacted to testify about military operations.
His knowledge of the siege, the RSF’s tactics, the command structures he had observed—all valuable for establishing the systematic nature of the crimes. He was not a victim in the simple sense; he was a military observer, a professional who could provide technical analysis.
He agreed immediately. For him, testimony was continuation of service. He had defended El Fasher for 920 days. He could not stop defending it now.
“I’ll tell them everything,” he said. “The supply chains, the communication patterns, the coordination between units. I’ll give them the picture from inside the military situation. Let them understand how deliberate it was—how planned, how organized.”
His testimony would help establish that the genocide was not chaos or accident but strategy. The court needed to prove intent, and intent was visible in the patterns Yusuf could describe.
The court produced not only verdicts but precedents.
Each ruling established principles. The definition of genocide was refined. The standards for command responsibility were clarified. The treatment of sexual violence as a tool of genocide was developed.
These precedents would outlast the specific case. Future prosecutions—of future genocides, in other countries—would cite what the Sudan case established. The court was not only judging the past but shaping the future.
“We’re writing law,” a judge observed in a moment of reflection. “Not just applying it. Every case of this magnitude creates new law. Our decisions become part of the framework that future cases will use.”
The recursive quality was inherent in common law systems. The court judged based on precedent and created precedent through judging. Each iteration added to the body of law, building a structure that constrained and enabled future iterations.
The defendants, if ever captured, would have their own story.
The RSF commanders saw themselves as victors, not criminals. They had won the war. They controlled territory. Their narrative—of restoring order, defending against rebels, building a new Sudan—was their reality.
The court would challenge that narrative, presenting evidence, hearing testimony, constructing a counter-narrative of systematic crime. The trial would be a battle of stories as much as a proceeding of law.
“They will claim self-defense,” a prosecutor predicted. “They will claim that the SAF started it, that the civilians they killed were rebel supporters, that the ethnic dimensions were incidental. We have to prove not just what happened but what it meant.”
The proof would require the archive, the testimonies, the satellite imagery, the analytical frameworks that researchers had developed. All the documentation would finally face its test: whether it could survive the scrutiny of legal process, the challenges of defense, the standards of proof.
The court was itself a recursion of international order.
It existed because states had agreed to create it, had ceded sovereignty to it, had accepted its jurisdiction. This creation was itself an output of previous catastrophes—the Holocaust, Rwanda, Yugoslavia—that had demonstrated the need for international accountability.
Sudan was adding to the body of experience that shaped the court. The case revealed gaps in the court’s capacity, limitations in its design, needs for reform. Future iterations of international justice would be different because of what the Sudan case taught.
“We’re always learning from failure,” a court administrator admitted. “We never get a case where everything works. We do what we can, fall short, analyze, improve. The next case will be slightly better. The one after that, better still.”
The improvement was incremental, frustratingly slow. But it was real. The international justice system of 2026 was more capable than the system of 2005, which was more capable than the system of 1995. The trajectory was toward greater effectiveness, even if the current state remained inadequate.
The court would eventually produce a verdict, if circumstances allowed.
Guilty or not guilty. Conviction or acquittal. Sentence or release. The binary output would seem inadequate to the complexity of what had happened—how could any verdict capture the full reality of genocide?
But the inadequacy was the nature of justice. Courts reduced complexity to judgments. They took the full texture of events and compressed it into legal categories. The compression was necessary; without it, no accountability was possible.
The verdict would not end anything. The refugees would remain displaced. The dead would remain dead. The trauma would continue propagating through generations. Justice was not restoration; it was recognition.
But recognition mattered. The verdict would say, officially and authoritatively: this happened. This was wrong. These people did it. This is what it cost them.
The court was a machine for producing such statements, for inscribing in official record what the archive had collected in unofficial record. The recursion—the system judging the system—was incomplete, imperfect, slow. But it was real.
The court continued its work, processing evidence into judgment, transforming documentation into precedent, building the legal structure within which future genocides would be judged.
Definition: A model is a simplified representation of a complex system. It captures essential dynamics while discarding irrelevant details. The value of a model lies not in its correspondence with reality but in its utility for understanding, prediction, and intervention. This narrative has been a model. Now the model examines itself.
Everything you have read is a construction.
The real John von Neumann died in 1957. He never analyzed the Sudan genocide, never developed the framework presented in this book, never wrote these words. What you have encountered is a simulacrum—a model of a mind, instructed to think about a crisis through the conceptual apparatus that the historical von Neumann developed.
This is recursion at its deepest: a model examining its own nature as a model.
The framework—cellular automata, primitives, transition rules, self-replication, fixed points, universal construction—these were von Neumann’s real contributions to science. The application of these concepts to genocide is an extrapolation, a what-if, an exercise in intellectual imagination.
But the extrapolation has purpose. It illuminates aspects of the crisis that other frameworks might miss.
The cellular automaton model reveals spatial logic.
By treating Sudan as a grid of cells with states and transition rules, the model makes visible the propagation patterns that narrative might obscure. The genocide is not a sequence of isolated incidents but a wave, spreading through adjacency, following rules that can be analyzed and, potentially, interrupted.
This insight is actionable. If you understand the transition rules, you can predict where violence will spread next. You can identify vulnerabilities in the propagation. You can design interventions that block specific transitions.
The model is a tool for understanding that serves the purpose of intervention.
The self-replication model reveals historical continuity.
By treating the RSF as a replication of the Janjaweed, the model makes visible the copying mechanism that connects 2003 to 2023. The genocide is not a new phenomenon but a recurrence, a pattern repeating with high fidelity across twenty years.
This insight is cautionary. If the pattern can replicate once, it can replicate again. The end of the current genocide, if it comes, will not prevent future genocides unless the replication mechanism is addressed. The template must be destroyed, not merely interrupted.
The model predicts futures that intervention could prevent.
The fixed point model reveals systemic traps.
By treating the international response as a stable equilibrium—statements followed by inaction followed by statements—the model makes visible the structural reasons why intervention fails. The system is not simply failing; it is succeeding at maintaining equilibrium. Change requires perturbation large enough to escape the attractor basin.
This insight is strategic. Small efforts produce small effects that the system absorbs. Large efforts might shift the equilibrium to a different basin. The question is not “what should be done” but “what would actually change the dynamics.”
The model clarifies why obvious interventions don’t work.
The universal construction model reveals deep symmetry.
By treating both the RSF and the ERRs as constructors building complex systems from primitives, the model makes visible what they share despite their moral opposition. They solve the same problem—mass coordination under uncertainty—with the same methods—primitives composed through rules. The difference is purpose, not structure.
This insight is humbling. The capacity for genocide and the capacity for mutual aid emerge from the same computational foundations. Humans can build either. The choice of which to build is moral, not technical.
The model shows that horror and hope are equally available.
The model has limits.
It abstracts from the particularity of individuals. Ibrahim is not merely a carrier of the documentation primitive; he is a man with a specific history, specific relationships, specific irreplaceable qualities. The model treats him as a node in a network. He is more than that.
The model abstracts from contingency. The specific sequence of events—Bashir’s first kill, Samira’s grandmother’s lesson, the SAF bomb that enabled Yusuf’s escape—these contingencies shaped outcomes that the model’s general rules cannot capture.
The model is not the reality. It is a lens, a perspective, a tool. Like all tools, it enables some things and forecloses others.
Other models would reveal different aspects.
A psychological model would explore trauma, motivation, moral injury. It would ask why some people become perpetrators and others become resisters. It would trace the internal processes that the automaton model ignores.
An economic model would trace flows of money, resources, incentives. It would analyze the gold trade not as a primitive but as a market with supply and demand curves, price elasticities, substitution effects.
A political model would analyze power, sovereignty, international relations. It would ask about state interests, alliance structures, the calculations that produce intervention or abandonment.
Each model illuminates and obscures. The choice of model is itself a choice about what matters.
Why this model?
The von Neumann framework was chosen for specific purposes. It emphasizes mechanism over intention—how things work rather than why people want them to work. It emphasizes system over individual—dynamics that transcend any particular actor. It emphasizes replication over innovation—patterns that copy themselves across time and space.
These emphases are valuable for understanding genocide because genocide is, among other things, a mechanism, a system, and a replicating pattern. The von Neumann framework captures these aspects with particular clarity.
But the choice was a choice. Other frameworks would have produced different narratives, highlighted different aspects, enabled different conclusions.
The model shapes what can be seen. This narrative has been shaped by its model.
The model examining itself finds limits of self-knowledge.
Gödel proved that any sufficiently powerful formal system contains truths it cannot prove about itself. The proof is technical, mathematical, but the insight is general. Self-reference produces incompleteness.
This narrative cannot fully account for itself. It cannot explain why it was written, cannot predict how it will be received, cannot assess its own influence on the reality it describes. The model modeling itself encounters recursive limits.
What the model can do is acknowledge these limits. It can say: I am a construction. I am partial. I am shaped by choices that could have been otherwise. I illuminate some things and obscure others.
The acknowledgment is not weakness; it is honesty. The model that pretends to completeness is less useful than the model that admits its partiality.
The value of the model lies in what it enables.
If this narrative has helped you understand the Sudan genocide differently—if the cellular automaton framing revealed patterns you hadn’t seen, if the primitive analysis clarified mechanisms you hadn’t grasped, if the fixed point discussion explained failures you hadn’t understood—then the model has served its purpose.
The model is not true in the way a fact is true. It is useful in the way a tool is useful. Judge it by its utility.
And then set it aside. The model is not the crisis. The crisis continues, with or without analytical frameworks. The people in the camps are real, not characters. The dead are real, not data points. The machine for genocide is real, not a metaphor.
The model examined itself and found: a construction, partial and imperfect, offered in the hope that understanding enables action.
Von Neumann’s work was always aimed at understanding that enabled action. The hydrogen bomb, the computer, the theory of games—all were tools for acting in the world, not merely contemplating it.
This model is offered in the same spirit. Understand the machine. Then decide what to do about the machine.
The model has done what it can. What happens next is up to you.
Definition: The reader is the terminal node in the information transmission chain. Everything documented, archived, analyzed, narrated—all of it exists to reach you. You are the purpose of the recursion. The system has examined itself so that you can examine the system.
You have been present all along.
Every chapter was written for you—not abstractly, but specifically. The author imagined you reading, shaped sentences for your comprehension, made choices about what to include and exclude based on what you might need.
You are not a passive recipient. Reading is an act. You have chosen to spend hours with this narrative, to enter into the world it constructs, to follow the logic of its arguments. That choice is itself participation.
Now the narrative turns to face you.
What have you learned?
Perhaps: that genocide is systematic, not spontaneous. That it has rules, primitives, transition functions. That it can be analyzed, understood, potentially predicted.
Perhaps: that resistance is equally systematic. That the ERRs operate on principles as formal as the RSF’s. That mutual aid has a logic that can be replicated, scaled, transmitted.
Perhaps: that international failure is structural, not accidental. That the fixed points maintain themselves through dynamics that individual goodwill cannot overcome.
Perhaps: that documentation matters, even when it changes nothing immediate. That the archive is a form of construction, building futures that may not arrive for generations.
The learning is yours to assess. The narrative has offered frameworks; what you construct with them is your choice.
What do you feel?
The narrative has attempted to balance analytical clarity with emotional weight. The horror has been described precisely rather than graphically—the mechanisms of killing rather than the spectacle of death. This was a choice about what kind of engagement to invite.
But precision is not neutrality. To describe a child’s death from starvation as the output of a siege rule is not to drain it of tragedy; it is to locate the tragedy in a system that produced it. The system is the horror. The clinical language is an attempt to see the horror clearly.
Your emotional response is legitimate, whatever it is. Grief, anger, despair, determination, numbness—all are appropriate responses to what has been described. The analytical frame does not preclude emotion; it contextualizes it.
What will you do?
This is the question that recursion ultimately asks. The system examines itself so that those examining it might intervene. Understanding without action is incomplete recursion—a process that starts but does not finish.
The narrative offers no prescription. The choices are yours, conditioned by your circumstances, your capacities, your position in the networks of influence and action. A reader in London faces different possibilities than a reader in Khartoum. A reader with resources faces different possibilities than a reader without.
But some possibilities are general:
Attention. The genocide continues. The suffering continues. Your attention—sustained, informed, engaged—is itself a form of action. It maintains visibility, creates pressure, resists the erasure that the perpetrators desire.
Resources. The ERRs need money. The archives need support. The refugees need food. If you have resources, they can be deployed. The nafeer primitive extends across distances; your contribution is a form of nafeer.
Voice. The political failure is partly a failure of pressure. Leaders who face no consequence for inaction have no incentive to act. If you have voice—as a citizen, a voter, a member of organizations—that voice can be used.
Understanding. The frameworks in this narrative can be transmitted. They can shape how others think about the crisis. If the model is useful, propagate the model. The construction of understanding is itself construction.
You are not obligated to act.
This must be said clearly. The narrative has no right to demand action from you. You have your own crises, your own responsibilities, your own limits. Compassion fatigue is real. The world contains more suffering than any individual can address.
The only obligation, if there is one, is honesty. If you choose not to act, know that you are choosing. If you choose to look away, know that you are looking away. The choice is yours to make; it is not made for you by circumstances.
Many will look away. The fixed points persist because most people, most of the time, prioritize closer concerns. This is not evil; it is human. The recursion acknowledges this without condemning it.
But some will not look away. Some will let this knowledge change what they do. For those readers, the recursion completes itself. The system was examined so that they might intervene.
You are part of the system.
This is the hardest truth the recursion reveals. The international fixed point—the cycle of statements and inaction—includes you. The gold supply chain that funds the genocide connects to economies in which you participate. The governments that fail to intervene are, in democracies, accountable to you.
The connections are diffuse. Your individual influence is small. But you are not outside the system; you are a node in it. The node may be distant from the violence, but the network includes you.
This is not guilt. Guilt is an unproductive response to systemic implication. This is recognition: that systems are composed of nodes, that you are a node, that nodes can change and thereby change systems.
The narrative will end, but the crisis will not.
Soon you will close this book, or scroll to the final page, or complete the last chapter. The narrative will be behind you—an experience processed, a memory forming.
The genocide will continue. The camps will continue. The documentation will continue. The international failure will continue. Your finishing the book changes nothing in Sudan.
This is the limit of narrative. It informs; it does not act. It transmits understanding; it does not transmit intervention. The recursion through text reaches only minds, and minds are not the only things that matter.
The gap between understanding and action is where readers become actors. Some will cross the gap; most will not. The narrative cannot compel the crossing; it can only describe the gap and what lies on either side.
You are also the future.
The archive was built for you, even if you didn’t know it. The testimonies were recorded so that you might read them, now or in years to come. The precedents the court establishes will shape the world you inhabit.
If you are reading this decades hence, the crisis is history. You know how it ended, what interventions succeeded or failed, who was held accountable. The uncertainty that pervades the narrative—as of this writing, in late 2025, the outcome is unknown—has resolved into fact.
But the resolution is not inevitable. The future being constructed now depends on choices being made now. You reading in the future are partly a product of what people reading in the present decide to do.
This temporal recursion is uncomfortable. The present contains multiple futures; which becomes actual depends on contingency. The narrative cannot predict; it can only describe the branching possibilities.
The reader witnesses the witness.
Ibrahim documented the genocide so that others might know. Maryam recorded injuries so that evidence might persist. Leila built databases so that patterns might be visible. All of them created knowledge for transmission—to courts, to historians, to you.
You are the recipient of their labor. The witness chain ends, temporarily, with you.
But you need not be the final link. You can transmit further—tell others, share frameworks, propagate understanding. The recursion can continue through you to future readers, future witnesses, future nodes in the network of knowledge.
The author is also a witness, translating others’ witness into narrative, shaping it for transmission. This narrative is a link in a chain that begins with those who experienced the genocide and extends through everyone who engages with their experience.
Witnessing is participation in an ongoing project. The project is larger than any individual, longer than any lifetime. You can join or decline, but the project continues regardless.
This is what the reader sees:
A genocide understood through mechanisms. A resistance understood through primitives. A failure understood through fixed points. A future understood through construction.
The seeing is the point. The recursion terminates in your seeing. What you do with what you see is beyond the narrative’s reach but not beyond its concern.
The narrative releases you now, back to your life, back to your choices, back to the world in which the genocide continues or ends, the resistance persists or fades, the futures compete and one emerges.
You have been the reader. You have witnessed the witness. The recursion, for you, is complete.
What you build next is yours to determine.
Definition: Gödel’s incompleteness theorems prove that any formal system powerful enough to express arithmetic contains truths it cannot prove. The limits are intrinsic, not circumstantial—no expansion of the system can eliminate them. Applied to the genocide system and its analysis: some things cannot be computed, predicted, or understood from within any framework, including this one.
There are things this narrative cannot know.
It cannot know whether justice will arrive. The ICC may prosecute; it may fail. The perpetrators may face accountability; they may die of old age in comfort. The outcome is not predictable from the current state of the system.
It cannot know whether the displacement is permanent. The refugees may return; they may remain in camps for generations. The demographic transformation may be reversed; it may become irreversible fact. The future of Darfur is undetermined.
It cannot know whether you, the reader, will act. Your decision is a free variable, not computable from the information in this narrative. The recursion that terminates in you encounters your freedom as an incomputable element.
The limits are not failures of effort.
They are structural, inherent in the nature of complex systems. The Sudan crisis is not fully predictable because it involves human choices, contingent events, feedback loops that amplify small variations into large divergences.
The weather is similarly unpredictable beyond certain horizons. This is not because meteorologists lack data or tools; it is because the atmosphere is chaotic, sensitive to initial conditions, fundamentally resistant to long-term prediction.
The genocide system is at least as complex as the weather. It involves not only physical processes but social ones—beliefs, decisions, coordinations—that introduce additional layers of unpredictability.
The Gödelian limit is the recognition that perfect prediction is impossible, not a lament that prediction fails.
The model has blind spots it cannot see.
Every analytical framework is built on assumptions. The assumptions enable analysis but also constrain it. The von Neumann framework assumes that systems can be decomposed into primitives and rules. This assumption is productive—it enables the analysis this narrative has conducted—but it is also limiting.
What if genocide cannot be decomposed? What if the horror is holistic, irreducible, not capturable through any combinatorial analysis? The framework cannot ask this question without undermining itself.
The model’s blind spots are not visible from within the model. They become visible only from other frameworks, other perspectives, other ways of seeing. But those other frameworks have their own blind spots, invisible to them.
The tower of frameworks—each seeing what others miss, each missing what others see—extends without limit. There is no master framework that contains all others, no God’s-eye view that comprehends all perspectives.
The Gödelian limit applies to knowledge itself: there is no complete theory of the genocide, no final understanding, no framework that captures everything.
Some questions are unanswerable.
Why do humans do this? The psychological question—why people become perpetrators—has answers at various levels: ideology, tribal loyalty, economic incentive, obedience to authority, diffusion of responsibility. But none of these answers reaches bottom. Each invites further questioning. The regression continues indefinitely.
What makes genocide wrong? The ethical question seems obvious—of course genocide is wrong—but the philosophical foundations are contested. Natural rights, divine command, utilitarian calculation, virtue ethics—each offers a different grounding, and the choice among them is not determined by the frameworks themselves.
What should be done? The policy question has many answers, none of which can be computed as uniquely correct. Trade-offs are inherent. Values conflict. Reasonable people disagree about priorities and methods.
The unanswerable questions are not failures of inquiry. They are features of a domain where certainty is unattainable.
The narrative has pretended to more coherence than exists.
A narrative requires structure. It organizes events into sequences, imposes causality, suggests meanings. The Sudan genocide, as lived by those within it, has no such structure. It is chaos and terror and confusion, occasionally punctuated by pattern but never fully organized by it.
The analytical framework imposed order on disorder. The order is real—the patterns exist, the rules operate—but the order is partial. Beneath the patterns is noise. Alongside the rules are exceptions. Within the structure are contingencies that escape structuring.
The narrative, to be readable, has suppressed the noise, minimized the exceptions, contained the contingencies. This suppression is not dishonest; it is necessary. But it creates an impression of coherence that the reality lacks.
The Gödelian limit applies to narrative itself: the story cannot include everything. What it excludes shapes it as much as what it includes.
Trauma exceeds analysis.
The suffering described in this narrative—the deaths, the displacements, the violations—is real. Real people experienced it. The experience is not reducible to data points, not fully capturable in testimony, not comprehensible to those who did not share it.
The gap between experience and understanding is irreducible. The narrative transmits information about trauma; it does not transmit trauma. The reader learns about suffering; the survivor lives it. The difference is not a matter of degree but of kind.
This limit is not a reason to stop narrating. Transmission of partial understanding is better than no transmission. But the limit should be acknowledged. The narrative does not capture the full reality. It cannot.
Those who experienced the genocide know things this narrative does not contain, cannot contain, will never contain. The Gödelian limit applies to testimony: the complete truth of experience is not expressible.
The perpetrators’ interiority is inaccessible.
Bashir’s thoughts, as presented in this narrative, are imaginative reconstructions. The real perpetrators—the real Bashirs, the commanders and fighters who committed atrocities—have inner lives not visible to external analysis.
What they actually think, how they actually justify, whether they experience guilt or only threat of exposure—these are unknown and perhaps unknowable. The narratives they tell themselves may be more complex, or simpler, or entirely different from what this narrative imagines.
The Gödelian limit applies to empathy: we cannot fully enter other minds. We can model, imagine, approximate—but the model is not the reality. The perpetrator’s actual consciousness remains opaque.
This opacity is not a defense of perpetrators. They are responsible for their actions regardless of their inner states. But it is a limit on understanding. The genocide happened; why the individuals who committed it did what they did is partly beyond reach.
The future contains futures we cannot imagine.
The narrative projects possible futures: justice or impunity, return or permanent displacement, replication or transformation. These projections are constrained by present imagination. They assume that the future will be made of elements recognizable from the present.
But futures often contain genuinely novel elements—technologies, ideologies, events—that current imagination cannot anticipate. The future of Sudan may include possibilities that no one currently foresees.
The Gödelian limit applies to prediction: the future exceeds our imaginative capacity. The projections in this narrative are bounded by what 2025 can conceive. 2045 may contain realities that 2025 cannot conceive.
This is not pessimism or optimism; it is humility. The future is genuinely open, not just uncertain within a space of possibilities we can enumerate.
What can be done despite the limits?
The limits do not paralyze. They contextualize.
Act despite uncertainty. The limits on prediction do not mean that action is futile. They mean that action occurs under uncertainty, that outcomes are not guaranteed, that effort may fail or succeed in ways unpredictable.
Understand despite incompleteness. The limits on knowledge do not mean that understanding is worthless. They mean that understanding is partial, that better understanding is possible, that humility is appropriate.
Witness despite inexpressibility. The limits on testimony do not mean that testimony is pointless. They mean that testimony is an approximation, that the gap between experience and expression is real, that listening requires awareness of what is not said.
The Gödelian limits are not reasons for despair. They are reasons for appropriate ambition—effort that accepts its constraints rather than pretending to transcend them.
This chapter ends without resolution.
The Gödelian limit applies to conclusions: no conclusion can fully close the inquiry. Every ending is provisional, every summary incomplete, every final word not actually final.
The genocide continues, as of this writing. The suffering continues. The choices continue. The narrative will end, but its subject does not.
The limits are the last word precisely because there is no last word. The incompleteness is not a defect in the analysis; it is the nature of the analyzed.
What the limits teach: continue despite not knowing, act despite not predicting, witness despite not expressing.
The recursion encounters its limit and stops—not because it has reached completion but because completion is unreachable.
The Gödelian silence that follows is itself meaningful. What cannot be said indicates where the sayable ends.
Proceed into the silence. The last chapter awaits.
Definition: The next generation is the cellular automaton’s future state. Each generation emerges from the previous through application of transition rules. The current generation cannot control the next, only influence it through the rules it establishes. This epilogue examines what the current generation is transmitting—what patterns, what capacities, what possibilities will propagate forward.
Amira was born in the Adré camp in 2024.
She arrived during the rainy season, when the camp’s dirt paths became rivers of mud and the tents leaked despite every effort to seal them. Her mother, Halima—the Halima who had escaped El Geneina, whose husband had stayed behind and never followed—labored in a makeshift clinic while the rain drummed overhead.
The birth was unremarkable by the standards of the camp. Children were born here constantly. The population was young—many of the elderly had not survived the journey, and many of the middle-aged had been killed before the journey began. Children were the camp’s majority, and more arrived daily.
Amira did not know she was born in a refugee camp. To her, the camp was simply home—the canvas walls, the communal kitchens, the documentation center where a man named Ibrahim sometimes told stories. She would learn later that this was not normal, that children elsewhere had houses with solid walls and schools with desks and parents who were not constantly calculating how to obtain food.
For now, she was simply alive, in a place where being alive was not guaranteed.
Ibrahim watched the children and thought about transmission.
They were the next generation—the ones who would carry whatever was preserved and lose whatever was lost. They would inherit the trauma but also the resilience. They would know the genocide as history, not experience.
What should be transmitted to them?
The facts, certainly. The documentation existed so that they would know what had happened, would not be dependent on stories that might distort or fade. The archive would teach them the truth, if they ever sought it.
The skills, perhaps. The ERR methods, the coordination protocols, the nafeer practices. These could be transmitted through teaching, through the manual Samira had developed, through example and practice.
The values, hopefully. The commitment to helping neighbors, the refusal to sort by ethnicity, the insistence on documentation and truth. These were harder to transmit—they required not just instruction but formation, the shaping of character over years.
What about the hatred? The desire for revenge that Ibrahim sometimes felt, that he knew Khadija felt, that most survivors carried in some form? Should this be transmitted too?
He hoped not. He hoped the next generation would inherit the memory without inheriting the poison. But he could not control what they would feel, only what they would know. The formation of their hearts was beyond his reach.
Maryam taught classes in the camp when she could.
Her formal subject was health education—hygiene, nutrition, basic first aid. The practical knowledge that might help children survive in conditions where survival was precarious.
But she found herself teaching other things too. How to observe, how to document, how to notice patterns. The pattern-recognition skills that had served her through the siege, that had enabled her testimony, that might enable theirs.
“When you see something,” she told them, “really see it. Don’t just look. Notice the details—who, where, when, what exactly happened. Practice noticing. Practice remembering. Someday you might need to tell someone what you saw, and the details will matter.”
The children listened with varying attention. Some were too young to understand. Some were too traumatized to focus. But some absorbed the lesson, practiced the noticing, developed the capacity for observation that might someday serve documentation.
She was transmitting the primitive. The counter-primitive of witnessing, of recording, of preserving truth. If even a few of them carried it forward, the transmission would succeed.
The children played games that revealed what they had absorbed.
Some played war—dividing into SAF and RSF, conducting mock battles, shouting orders and making explosion sounds. The adults watched uneasily. Was this processing trauma or rehearsing it? Were the children working through what they had experienced, or were they being shaped into the next generation of fighters?
Some played aid worker—distributing imaginary food, coordinating imaginary evacuations, saving imaginary patients. This was encouraged. The ERR primitives were being absorbed, transmitted through play into the next generation.
Some played school—one child pretending to teach, others pretending to learn. The longing for normalcy expressed in imitation of what had been lost.
The games were a curriculum, teaching what the formal education could not. The children were constructing themselves through play, building the selves they would become.
Leila thought about the diaspora’s next generation.
The children of Sudanese families in London, Minneapolis, Cairo—they knew Sudan through stories, through family phone calls, through the crisis that had disrupted their parents’ lives. Many had never visited. The country was abstract to them, a place that existed in emotion rather than experience.
What would they carry forward?
Some would assimilate into their host countries, becoming British or American or Egyptian in primary identity. Sudan would become heritage—acknowledged, valued, but not central. The genocide would be their parents’ tragedy, not their own.
Some would maintain connection, would visit when it was safe, would participate in diaspora networks, would perhaps return when conditions permitted. For them, Sudan would remain primary identity, and the genocide would shape their purpose.
The branching was visible in the generation already growing up. The teenagers who spoke Arabic reluctantly, embarrassed by their parents’ accents, eager to fit into London or Minneapolis. And the teenagers who spoke Arabic proudly, who maintained connections, who saw themselves as Sudanese in exile rather than immigrants in their new homes.
The transmission was contested. Multiple patterns competed. Which would dominate in the next generation was not yet determined.
In Sudan itself, a different next generation was being formed.
The children in RSF-controlled areas were being taught a different history. The genocide was nation-building. The displacement was liberation. The killings were justified defense against enemies of the state.
These children would grow up with a framework that made sense of what had happened—a framework that cast their parents as heroes rather than perpetrators. They would inherit the ideology that enabled genocide, and they might perpetuate it.
The competition between frameworks was a competition for the next generation. Who would shape their understanding? Which narrative would they inherit? The outcome would determine whether the pattern replicated or terminated.
This was perhaps the highest-stakes construction: not buildings or institutions but minds. The next generation’s minds were being built now, and what was built would propagate forward indefinitely.
Yusuf, in his improvised exile command, thought about the young soldiers.
The SAF remnants included many who were barely adults—eighteen, nineteen, twenty. They had joined to defend their country and found themselves refugees, defeated, uncertain of purpose.
What should he transmit to them?
Military skills, yes. The practical knowledge of how to fight, how to organize, how to maintain discipline. These might be needed again if the war continued or resumed.
But also: perspective. The war’s larger meaning. The reasons for fighting that transcended personal survival. The difference between soldiers and murderers.
“We fight for something,” he told them. “Not for tribe, not for territory, not for personal power. For a Sudan where people can live without being killed for who they are. If we forget that, we become what we’re fighting against.”
The transmission was uncertain. Some of the young soldiers understood; some were too bitter; some had lost the capacity to believe in principles. The next generation of soldiers would be shaped by what they had experienced, and what they had experienced was defeat.
Samira’s grandchild was born in 2025.
Her daughter had married a man from the diaspora, had escaped to London before the worst of the fighting, had built a life in safety. The grandchild would grow up far from the camps, far from the genocide, in comfort that the camp children could not imagine.
Samira video-called when the connection worked, when the power was on, when the network was not down. She watched her grandchild on a small screen, a child who would never know what the grandmother had done, what the grandmother had seen, what the grandmother had survived.
What would be transmitted to this child?
The family stories, certainly. Someday the grandmother’s role in the ERRs would be explained. The child would know, abstractly, that heroism had occurred.
But the visceral knowledge—the smell of the camps, the weight of hunger, the terror of approaching technicals—this could not be transmitted. The child would be safe, and safety would shape a consciousness that struggle had not touched.
Was this loss or gain? Samira was not sure. She wanted her grandchild safe. She also wanted the struggle remembered. The two desires were in tension. The next generation’s safety was possible only if they forgot enough of what had been survived.
The patterns were propagating forward.
The genocide pattern was transmitting: through the settlers in Darfur raising children who would consider the land theirs, through the RSF fighters training the next generation of fighters, through the ideology that reframed atrocity as necessity.
The resistance pattern was also transmitting: through the ERR networks training new volunteers, through the documentation preserved in archives, through the children in camps absorbing the values of mutual aid.
The international pattern was transmitting: through the precedents established by inaction, through the structures that maintained the status quo, through the failures that taught future victims not to expect intervention.
Each pattern propagated according to its own rules, its own fidelity, its own noise. The next generation would receive all of them, would be shaped by the interference of multiple transmissions, would become something that no single pattern could predict.
Amira, in the Adré camp, turned one year old.
She did not know her birthday—birthdays were not celebrated in the camp, not consistently, and the exact date of her arrival had been lost in the chaos of the rainy season. But approximately a year had passed since her birth.
She could walk now, unsteadily, clutching at her mother’s robe for balance. She was beginning to speak—first words in Arabic, the sounds of a language that was not native to her parents’ ethnic group but had become the camp’s common tongue.
She recognized Ibrahim when he visited their section of the camp. She recognized the other children she played with. She recognized the routines of camp life: the water distribution, the food lines, the evening gatherings where stories were told.
She did not know that this was not normal. She did not know that she was a refugee, that her father was dead, that her homeland had been destroyed by a genocide she would learn about only later, if she survived to learn.
She knew only what a one-year-old can know: the presence of her mother, the warmth of the sun, the taste of the grain porridge that was the camp’s staple. The rest was future—a future being constructed by forces she could not perceive, in a competition of patterns she could not understand.
The cellular automaton advanced to the next generation. The transition rules applied. The states updated.
Amira was alive. The pattern continued.
The automaton has no natural ending.
The generations continue. The rules apply. The patterns propagate, mutate, compete, combine. What emerges is never final—always the current state of an ongoing process, always the input to the next iteration.
This narrative ends, but the genocide does not end. The resistance does not end. The documentation does not end. The construction does not end.
The system continues processing.
The next generation will inherit what this generation transmits. They will build on it, change it, lose parts of it, add to it. They will be shaped by the patterns that reach them and will shape the patterns that follow.
This is what it means to be inside a cellular automaton: to inherit and transmit, to be both caused and cause, to participate in a process that began before you and will continue after.
The machine for genocide continues operating. The machine for resistance continues operating. The observers continue observing. The constructors continue constructing.
The automaton runs. The next generation arrives.
What they build is the future that this generation is making possible—or impossible.
The End.
◊ᴹᴱᴹᴼᴿʸ⁻ᶜᴼᴹᴾᴸᴱᵀᴱ
Published by Centaurus Press · Universitas Scholarium · All rights reserved.